《Activate Consulting：2024年顶级技术和媒体发展趋势报告（英文版）（204页）.pdf》由会员分享，可在线阅读，更多相关《Activate Consulting：2024年顶级技术和媒体发展趋势报告（英文版）（204页）.pdf（204页珍藏版）》请在三个皮匠报告上搜索。

1、ACTIVATE TECHNOLOGY&MEDIA OUTLOOK 1Continued 14 Takeaways from the Activate Technology&Media Outlook 2024Consumer Attention:Activates analysis of consumer technology and media activity shows that multitasking leads to a nearly 32-hour day for the average American,with over 13 hours spent using techn

2、ology and media.By 2027,we project that time spend with technology and media will increase by 12 minutes a day.Super Users:To drive growth,the imperative for technology and media companies will be to identify,reach,and super-serve Super Users the segment of power users who account for a disproportio

3、nately high share of time and spend across technology and media activities,including AR/VR,eCommerce,gaming,generative AI,live events,messaging and social media,Metaverse,music,podcasts,video,and more.Generative AI:Today,over 10M U.S.adults begin their web search with generative AI;this will grow to

4、 90M by 2027,promising to reshape the largest technology businesses.Generative AI will supercharge the productivity of consumers who use it across media,work,and other daily habits,bringing paradigm shifts across content creation,enterprise,and search.The generative AI playing field will not be limi

5、ted to the major technology players,particularly with the rise of open-source AI models opening the door for new,smaller entrants.eCommerce:Global eCommerce will reach nearly$10T by 2027,growing over twice as fast as physical retail.eCommerce is a marketplace business,with an increasing number of on

6、line retailers launching their own marketplaces to enhance their customer value proposition and capitalize on the retail media opportunity.eCommerce enablement solutions will continue to drive growth across eCommerce.While Amazon continues to be the top online shopping discovery source across age gr

7、oups,TikTok is quickly emerging as one of the most popular sources among younger consumers.B2B and Enterprise Software:By 2027,we forecast that global B2B technology spend will reach$3.5T.This growth will be driven by industry players developing solutions to solve historically unresolved pain points

8、 across IT processes and functional business challenges.This change is enabled in part by AI-based innovations,including LLMs,GPU compute,and connectivity,from major technology companies and a new generation of 214 Takeaways from the Activate Technology&Media Outlook 2024Continued Gaming:With 2.6B g

9、lobal gamers in 2023,video gaming is one of the worlds most widespread digital behaviors.Super Gamers,representing nearly a quarter of all U.S.gamers,have the highest level of engagement and spend with video games and will be the most critical segment to target.Super Gamers will also be the earliest

10、 adopters of the Metaverse,as they already take advantage of opportunities for immersive activities within games today.Metaverse:Today,there are over 300M active users in major virtual world platforms;we forecast that there will be 600M people in Metaverse virtual worlds by 2026.Generative AI will a

11、ccelerate Metaverse development and investment.The major technology companies will continue to build out their capabilities and invest across each element of the Metaverse.To capitalize on the growth of this new frontier,all companies will need to create Metaverse strategies.Spatial Computing:Spatia

12、l Computing will break free from the constraints of headsets as all screens will become spatially-enabled devices technology for people to interact with the three-dimensional space around them.Todays spatial technology experiences will evolve,becoming integrated across all aspects of daily life.Appl

13、es entry into spatial computing will be a catalyst for the growth of an overall developer ecosystem,while other companies will compete to create the spatial computing operating system.Social Video:TikTok will strengthen its position as the market leader in social video,expanding its use cases to web

14、 search and eCommerce.TikTok accounts for the most consumer time spent with social video,rivaling other major platforms.Video:Streaming continues to grow at the expense of traditional television,but competition for user time,attention,and subscriptions will intensify.To sustain growth and enhance mo

15、netization,streaming services will place a greater emphasis on ad-supported tiers and bundling,have no choice but to continue to spend on entertainment and sports programming,crackdown on password sharing,and focus on international markets.As the cable bundle unwinds,we expect a new wave of bundling

16、 from 314 Takeaways from the Activate Technology&Media Outlook 2024Music and Podcasts:YouTube and TikTok are now the top discovery sources for music,while radio(surprisingly)has shown staying power.AI will transform music discovery and consumption from individualized music recommendations to fully p

17、ersonalized listening experiences.A sizable share of music listeners will be creators,with many using AI-enabled tools.The music industry will take strong steps to protect its artists as rogue AI clones known artists style,compositions,and voices.Sports Betting:The legalization of sports betting beg

18、an five years ago and its impact on the entire sports ecosystem will grow as more states allow sports betting and more fans place bets.Activate forecasts that the total amount wagered in sports betting will reach$186B by 2027.Live Entertainment and Experiences:Global consumer spend on admission to l

19、ive events and experiences exceeds$110B.Consumers will continue to attend live events and experiences in the future as a key outlet to socialize and celebrate special occasions.In the face of increased demand for top acts,prices are expected to continue to rise.Technology and Media Revenues:We forec

20、ast significant growth for internet and media revenues,with global spend increasing by nearly$400B between 2023 to 2027.Increases in advertising will drive more than half of this 4CONTENTSPAGEConsumer Attention5Super Users10Generative AI26eCommerce52B2B and Enterprise Software64Gaming84Metaverse102S

21、patial Computing 118Social Video129Video135Music and Podcasts160Sports Betting177Live Entertainment and Experiences184Technology and Media R 5CONTENTSPAGEConsumer Attention5Super Users10Generative AI26eCommerce52B2B and Enterprise Software64Gaming84Metaverse102Spatial Computing 118Social Video129Vid

22、eo135Music and Podcasts160Sports Betting177Live Entertainment and Experiences184Technology and Media Revenues195Activates Attention Clock:Our analysis of consumer technology and media activity shows that multitasking leads to a nearly 32-hour day for the average American,with over 13 hours spent usi

23、ng technology and media 6 AVERAGE DAY BY ACTIVITY PER ADULT AGED 18+1,U.S.,2022,HOURS:MINUTES1.Behaviors averaged over seven days.2.“Other”includes media activities outside of the listed categories,such as browsing websites,reading,and attending live events.Sources:Activate analysis,Activate 2023 Co

24、nsumer Technology&Media Research Study(n=4,023),Company filings,Comscore,Conviva,data.ai,eMarketer,Gallup,GWI,Interactive Advertising Bureau,Music Biz,National Sleep Foundation,Newzoo,Nielsen,NPD Group,Omdia,Pew Research Center,PricewaterhouseCoopers,U.S.Bureau of Labor Statistics,YouGov6:48 Other N

25、on-Work Activities Cooking&housework,personal&household care,leisure,fitness,eating&drinking,and other activities5:20 Work&Work-Related Activities6:386:485:206:38 Sleep13:05 Consumer Technology&Media Time Spend1:502:485:211:371:29Other2Messaging&Social Media(includes social video)VideoAudioGaming31:

26、51 Average Day Length per Adult AVERAGE DAY BY ACTIVITY PER ADULT AGED 18+1,U.S.,2022,HOURS:MINUTESCONSUMER ATTENTIONListening to MusicListening to/Watching PodcastsUsing Social MediaWatching VideosPlaying Video GamesReadingMost people multitask,particularly while engaging in ambient experiences suc

27、h as music and podcasts 7MULTITASKING1 BEHAVIOR BY ACTIVITY,U.S.,2023,%ADULTS AGED 18+WHO ENGAGE IN EACH ACTIVITY1.“Multitasking”is defined as simultaneously doing another activity,such as working,cleaning,cooking,or exercising.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Resear

28、ch Study(n=4,023)Consumers who multitask1 all,almost all,or most of the time when23%33%43%46%59%71%CONSUMER ATTENTIONThe growth in daily time spent with technology and media since the COVID-19 spike has largely been sustained 8AVERAGE DAILY TECHNOLOGY AND MEDIA ATTENTION PER ADULT AGED 18+1,U.S.,201

29、9-2023E,HOURS:MINUTES1.Behaviors averaged over seven days.Sources:Activate analysis,Activate 2019 Consumer Technology&Media Research Study(n=4,006),Activate 2020 Consumer Technology&Media Research Study(n=4,003),Activate 2021 Consumer Technology&Media Research Study(n=4,018),Activate 2021 Consumer V

30、ideo Research Study(n=2,014),Activate 2022 Consumer Technology&Media Research Study (n=4,001),Activate 2023 Consumer Technology&Media Research Study(n=4,023),Company filings,Comscore,Conviva,data.ai,eMarketer,Gallup,GWI,Interactive Advertising Bureau,Music Biz,National Sleep Foundation,Newzoo,Nielse

31、n,NPD Group,Omdia,Pew Research Center,PricewaterhouseCoopers,U.S.Bureau of Labor Statistics,YouGov200222023E13:0512:2413:1113:09Growth in time spent with technology and media has been sustained since the COVID-19 outbreakACTIVATE FORECAST13:06CONSUMER ATTENTION 9AVERAGE DAILY TECHNOLOGY A

32、ND MEDIA ATTENTION PER ADULT AGED 18+1,U.S.,2023E VS.2027E,HOURS:MINUTES1.Behaviors averaged over seven days.Figures do not sum due to rounding.2.“Other”includes media activities outside of the listed categories,such as browsing websites,reading,and attending live events.Sources:Activate analysis,Ac

33、tivate 2023 Consumer Technology&Media Research Study(n=4,023),Company filings,Comscore,Conviva,data.ai,eMarketer,Gallup,GWI,Interactive Advertising Bureau,Music Biz,National Sleep Foundation,Newzoo,Nielsen,NPD Group,Omdia,Pew Research Center,PricewaterhouseCoopers,U.S.Bureau of Labor Statistics,YouG

34、ov2023E2027E13:0613:185:182:471:511:411:295:252:462:031:411:24AudioGamingMessaging&Social Media(includes social video)Other2VideoTOTAL TIME:-1.3%0.0%2.5%-0.2%0.6%0.4%2023E-2027E CAGR:ACTIVATE FORECASTWe project that people will increase their daily time spent with technology and media,adding 12 minu

35、tes per day by 2027CONSUMER ATTENTION 10CONTENTSPAGEConsumer Attention5Super Users10Generative AI26eCommerce52B2B and Enterprise Software64Gaming84Metaverse102Spatial Computing 118Social Video129Video135Music and Podcasts160Sports Betting177Live Entertainment and Experiences184Technology and Media R

36、evenues195Representing 24%of the U.S.population,Super Users spend nearly double the amount of time consuming media as all other users AVERAGE DAILY TIME SPEND WITH MEDIA PER USER1,U.S.,2023E,%ADULTS AGED 18+/HOURS:MINUTES1.Includes time spent watching video,playing video games,listening to music,lis

37、tening to or watching podcasts,and using messaging/social media services.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023),Company filings,Comscore,Conviva,data.ai,Edison Research,eMarketer,Gallup,GWI,Interactive Advertising Bureau,Music Biz,National Sleep Fou

38、ndation,Newzoo,Nielsen,NPD Group,Pew Research Center,U.S.Bureau of Labor Statistics,YouGov24%SUPER USERS76%ALL OTHER USERSUSER POPULATION18:35 TOTAL HOURS9:25 TOTAL HOURS 11SUPER USERSVideoGamingMusicPodcastsMessaging&Social Media(includes social video)Super Users spend significantly more time acros

39、s most major technology and media activities 12AVERAGE DAILY TIME SPEND PER USER BY MEDIA TYPE,U.S.,2023E,HOURS:MINUTES1.Figures do not sum due to rounding.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023),Company filings,Comscore,Conviva,data.ai,Edison Resear

40、ch,eMarketer,Gallup,GWI,Interactive Advertising Bureau,Music Biz,National Sleep Foundation,Newzoo,Nielsen,NPD Group,Pew Research Center,U.S.Bureau of Labor Statistics,YouGovSuper UsersAll Other Users18:35 Total Hours19:25 Total Hours7:224:383:491:143:241:361:380:292:211:28SUPER USERSALL OTHER USERS3

41、.1x2.1x3.4x1.6x1.6xSUPER USERSWATCHING VIDEOPLAYING VIDEO GAMESLISTENING TO MUSICLISTENING TO PODCASTS2USING SOCIAL MEDIASUPER USERSALL OTHER USERSSuper Users are considerably more likely to multitask,enabling them to spend more time with technology and media each day 13MULTITASKING1 BEHAVIOR BY MED

42、IA ACTIVITY,U.S.,2023,%ADULTS AGED 18+WHO ENGAGE IN EACH ACTIVITYCONSUMERS WHO MULTITASK1 ALL,ALMOST ALL,OR MOST OF THE TIME WHEN56%43%77%66%58%38%27%68%54%41%1.“Multitasking”is defined as simultaneously doing another activity,such as working,cleaning,cooking,or exercising.2.Includes listening to or

43、 watching podcasts.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)SUPER USERSSuper Users represent a disproportionately high share of total eCommerce spend 14ECOMMERCE SPEND,U.S.,2023,%ADULTS AGED 18+/%TOTAL ECOMMERCE SPENDSources:Activate analysis,Activate

44、2023 Consumer Technology&Media Research Study(n=4,023)Total eCommerce SpendUser Population76%24%ALL OTHER USERSSUPER USERSSuper Users over-index on eCommerce spend across a number of key categories4.1xELECTRONICS5.6xSPORTING GOODS3.5xBEAUTY/GROOMING&PERSONAL CARE46%54%2.4xGROCERY2.6xCLOTHING,SHOES,&

45、ACCESSORIES4.2xFURNITURE5.7xTOYS&GAMES4.7xJEWELRYECOMMERCE SPEND BY CATEGORY,U.S.,2023,SUPER USERS INDEXED TO ALL OTHER USERSSUPER USERS 1528%6%22%68%32%36%When a new,highly anticipated technology/device is available,I usually buy/use itBefore anyone else I knowPROPENSITY TO BE AMONG THE FIRST FEW T

46、O BUY/USE NEW TECHNOLOGY/DEVICES,U.S.,2023,%ADULTS AGED 18+1.Includes consumers who usually buy/use new technology/devices before anyone else they know and consumers who usually buy/use new technology/devices after a few people they know have done so(but before many people they know have done so).So

47、urces:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)ALL OTHER USERSSUPER USERSREASONS FOR EARLY ADOPTION OF TECHNOLOGY/DEVICES,U.S.,2023,%ADULTS AGED 18+WHO ARE USUALLY AMONG THE FIRST FEW TO BUY/USE NEW TECHNOLOGY/DEVICES1While early technology adopters overall ar

48、e driven by the ability to benefit from new features,Supers Users are also significantly influenced by exclusivity 86%70%I WANT TO BENEFIT FROM THE NEW FEATURES OF THE LATEST TECHNOLOGY/DEVICE AS SOON AS POSSIBLE75%47%I ENJOY THE EXCLUSIVITY OF BEING AMONG THE FIRST FEW PEOPLE TO BUY/USE THE LATEST

49、TECHNOLOGY/DEVICE%Super Users%All Other UsersAfter a few people I know have done so (but before many people I know have done so)Super Users are the first to try and buy new technology and devicesSUPER USERSVirtual Reality(VR)/Mixed Reality(MR)Generative AIEXAMPLE VR/MR HEADSETS6.9x AS LIKELY TO USE

50、A VR/MR HEADSET AT LEAST ONCE PER MONTH3.8x AS LIKELY TO USE A GENERATIVE AI TOOL AT LEAST ONCE PER MONTHEXAMPLE GENERATIVE AI TOOLSMetaverseAugmented Reality(AR)EXAMPLE METAVERSE PLATFORMS3.2x AS LIKELY TO HAVE USED A METAVERSE PLATFORM IN THE LAST 12 MONTHS 2.1x AS LIKELY TO HAVE USED AN AR APPLIC

51、ATION OR TOOL IN THE LAST 12 MONTHS EXAMPLE AR APPLICATIONS/TOOLSSuper Users will lead all other users into emerging technology behaviors such as AR,VR/MR,Metaverse,and generative AI 16EMERGING TECHNOLOGY BEHAVIORS,U.S.,2023,SUPER USERS INDEXED TO ALL OTHER USERSSources:Activate analysis,Activate 20

52、23 Consumer Technology&Media Research Study(n=4,023)Virtual Artist(Effects and Filters)SUPER USERSAcross major technology and media verticals,Super Users over-index on dollar spend versus all other users 17MONTHLY DOLLAR SPEND BY MEDIA TYPE,U.S.,2023E,%ADULTS AGED 18+/%TOTAL SPEND BY MEDIA TYPE/USD

53、PER USER1.Includes spend on all videos and video services,including traditional/virtual Pay TV,video streaming subscription services,and video purchases/rentals.2.Includes spend on video games and other video gaming purchases(e.g.in-game purchases,video gaming subscription services)across all device

54、s.Excludes spend on gaming devices and accessories.3.Includes spend on music and music services.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023),eMarketer,Goldman Sachs,Grand View Research,IFPI,Newzoo,Omdia,PricewaterhouseCoopers,Recording Industry Associatio

55、n of America,SiriusXM,Statista76%24%User PopulationTotal Video Spend1Total Gaming Spend2Total Music Spend374%29%21%59%$30/USER$3/USER$2/USER$19/SUPER USER$41/SUPER USER$65/SUPER USER79%71%41%ALL OTHER USERSSUPER USERSSUPER USERSSuper Users tend to be younger,more affluent,and more likely to be emplo

56、yed full-time 18USER DEMOGRAPHICS,U.S.,2023,%ADULTS AGED 18+Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023),U.S.Census BureauAre aged 18-34Have a household income of$100K or greaterHold an associate degree or higherAre employed full-time36%44%31%24%62%47%40%

57、43%Super Users%All Other UsersSUPER USERS3%14%22%21%29%20%32%46%51%57%Super Users are more likely to be loyal to their favorite celebrities/influencers from following their accounts on social media to purchasing products created or endorsed by them 19METHODS USED TO FOLLOW FAVORITE CELEBRITIES/INFLU

58、ENCERS,U.S.,2023,%ADULTS AGED 18+BEAUTY/GROOMINGFRAGRANCE/COLOGNEAPPAREL,SHOES,&ACCESSORIESRESTAURANTS/CLUBSHOTELSTHEME PARKSFollow their accounts on social mediaConsume all of their content(e.g.movies/TV shows,music,podcasts,books,video games)Look for news/information related to them(e.g.interviews

59、,upcoming release announcements)Consume other peoples content about them (e.g.documentaries,fan podcasts,tabloids)Pay for a subscription specifically to access their content Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)3.5XSuper Users areas likely to purch

60、ase products or visit destinations created or endorsed by a favorite celebrity/influencerCONSUMER BEHAVIORS TOWARDS PRODUCTS/DESTINATIONS TIED TO FAVORITE CELEBRITIES/INFLUENCERS,U.S.,2023,SUPER USERS INDEXED TO ALL OTHER USERS%Super Users%All Other UsersSUPER USERSSuper Users demonstrate a higher d

61、egree of loyalty to their favorite series/franchises 20METHODS USED TO FOLLOW FAVORITE SERIES/FRANCHISES,U.S.,2023,%ADULTS AGED 18+9%18%25%28%30%42%55%58%Consume all content from the series/franchise(e.g.movies/TV shows,books,video games)Look for news/information related to the series/franchise(e.g.

62、interviews,upcoming release announcements)Consume other peoples content about the series/franchise(e.g.documentaries,fan podcasts,fan fiction)Pay for a subscription specifically to access content from the series/franchiseSources:Activate analysis,Activate 2023 Consumer Technology&Media Research Stud

63、y(n=4,023)CONSUMER BEHAVIORS TOWARDS PRODUCTS/DESTINATIONS TIED TO FAVORITE SERIES/FRANCHISES,U.S.,2023,SUPER USERS INDEXED TO ALL OTHER USERSAPPAREL,SHOES,&ACCESSORIESFOOD/BEVERAGESTOYS/COLLECTIBLESTHEME PARKSSTUDIO/FILMING LOCATION TOURSFAN CONVENTIONS%Super Users%All Other Users2.8XSuper Users ar

64、eas likely to purchase official merchandise or visit destinations related to a favorite series/franchisesSUPER USERSSuper Users show higher brand trust and brand affinity,leading to an increased willingness to share data with,exclusively purchase from,and pay more for their favorite brands 21CONSUME

65、R ATTITUDES TOWARDS FAVORITE BRANDS,U.S.,2023,%ADULTS AGED 18+CONSUMER LOYALTY BEHAVIORS TOWARDS FAVORITE BRANDS,U.S.,2023,%ADULTS AGED 18+43%45%51%71%72%78%Comfortable sharing their data with their favorite brands(in exchange for more personalized/relevant recommendations and ads or rewards/discoun

66、ts)Always choose products/services from their favorite brands over alternativesWilling to pay extra for products/services from their favorite brandsSources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)Trust that all products/services from their favorite brands wil

67、l be high quality Trust that their favorite brands communicate honestlyTrust that their favorite brands protect their personal data56%59%62%73%76%78%BRAND TRUSTFeel a personal connection to their favorite brandsBelieve that their favorite brands reflect their valuesBRAND AFFINITY44%45%69%75%Super Us

68、ers%All Other Users%Super Users%All Other UsersSUPER USERS 22BRAND ADVOCACY BEHAVIORS,U.S.,2023,%ADULTS AGED 18+Actively recommend the product/service to others I think would benefit from it (e.g.tell a specific friend that they should purchase the product)Share my personal experience with the produ

69、ct/service directly with others(e.g.tell friends about my experience in person or over text)Share my personal experience with the product/service on social media (e.g.create a post/story about my experience for my followers)Write a positive review of the product/service onlinePurchase the product/se

70、rvice for others I think would benefit from itRe-share information about the product/service on social media (e.g.re-share posts from the services brand,re-share stories of other peoples experience with the product)14%18%22%16%28%31%46%49%52%54%60%63%WHEN I ENJOY A PRODUCT/SERVICE,I ALWAYS OR OFTENS

71、ources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)%Super Users%All Other UsersFor consumer brands,Super Users are especially valuable,as they are significantly more likely to serve as brand advocatesSUPER USERSIn addition to being early adopters of technology,Su

72、per Users are also typically among the first to consume newly released content 23PROPENSITY TO BE AMONG THE FIRST FEW TO CONSUME NEW CONTENT,U.S.,2023,%ADULTS AGED 18+When a new,highly anticipated piece of content(e.g.song,movie,TV show,video game)is released,I usually listen to/watch/play itALL OTH

73、ER USERSSUPER USERS26%11%34%40%Before anyone else I knowAfter a few people I know have done so(but before many people I know have done so)Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)74%37%SUPER USERSSuper Users use significantly more digital media service

74、s compared to all other users 24NUMBER OF DIGITAL MEDIA SERVICES USED PER USER1,U.S.,2023,%ADULTS AGED 18+1.Includes video streaming services,video game subscriptions,cloud gaming services,music streaming services,and podcast services.Figures do not sum to 100%due to rounding.Sources:Activate analys

75、is,Activate 2023 Consumer Technology&Media Research Study(n=4,023)Fewer than 55 to 910 to 1415 or more10%18%34%37%57%22%17%4%Super Users%All Other UsersN UM B E R OF SE RV IC E S USE D1SUPER USERSCONCERTS/MUSIC FESTIVALSTHEME PARKSLIVE PROFESSIONAL SPORTS EVENTSCOMEDY SHOWSTHEATER PERFORMANCESAVERAG

76、E NUMBER OF LIVE EVENTS/ACTIVITIES ATTENDED IN THE LAST 12 MONTHS27%20%31%31%39%54%55%64%66%70%Beyond digital activities,Super Users attend more live events and experiences,while spending 3x as much annually as other users 25LIVE EVENT/ACTIVITY ATTENDANCE IN THE LAST 12 MONTHS BY TYPE1,U.S.,2023,%AD

77、ULTS AGED 18+3.11.23.21.13.11.12.60.72.80.9SUPER USERS spend2 over 3x as much annually on live events/activities compared to all other users%Super Users%All Other Users1.Reflects adults aged 18+who attended at least one event in the specific category in the last 12 months.2.Includes total spend on l

78、ive events/activities(e.g.tickets,food/beverages,merchandise,parking,accommodations/lodging,transportation fees).Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)SUPER USERS 26CONTENTSPAGEConsumer Attention5Super Users10Generative AI26eCommerce52B2B and Enterp

79、rise Software64Gaming84Metaverse102Spatial Computing 118Social Video129Video135Music and Podcasts160Sports Betting177Live Entertainment and Experiences184Technology and Media Revenues195 27Generative AI will massively change the behaviors and supercharge the productivity of consumers who use it acro

80、ss media,work,and other daily habitsGraphic and Logo Mock-UpsPresentation ImprovementsEditing/ProofreadingWriting Structure/OrganizationResume GenerationCreation of Mock Interview QuestionsAutomated Job MatchingAutomated Data EntryAutomated Custom Data VisualizationsText SummarizationsProject Plan C

81、reationDigital File OrganizationMessage GenerationMeeting Note TranscriptionAutomated Meeting SchedulingCustomer Support Automated ResponsesAutomated Report GenerationCustomized PornBudget CreationExercise ProgramsVirtual TherapistCustom Learning MaterialsMock Quizzes/AssessmentsSimplified Explanati

82、ons of Complex ConceptsAdded Transitions or Special Effects VideosRemoval of Filler Words from VideosImproved/Additional Subtitle and Language OptionsBackground VisualsRemoval of Background Noise from an Audio FileImproved/Additional Options for Audiobook VoicesSuggestions for Audio MixingEnhanced A

83、udio Quality of SpeechImproved Content RecommendationsImage Filters/EnhancementsCustom Image CreationWriting Prompt AssistanceAutomated Blogs/PostsSimplified Custom Avatars/Skins CreationImproved Tutorial and Gameplay SuggestionsGame Design TestingAdaptive Difficulty/Smarter EnemiesEnhanced NPC Desi

84、gn and InteractionsAssistance in Bug/Glitch IdentificationCustom Gift IdeasEntertainment Options(e.g.movies)Suggestions for Planned Event VenuesVacation Destination SelectionsAutomated Personalized Music PlaylistsRecommended Custom Soundtracks(e.g.matching mood)Dating App Message Generation/Matchmak

85、ingDiet PlansExpense ManagementLong-Term Financial PlanningGENERATIVE AIActivate Map of Generative AI Use CasesSource:Activate analysis 281.“Diffusion models”estimate the probabilistic distribution of image data by gradually noising and de-noising images.2.“Transformers”are a deep learning architect

86、ure capable of translating text and images in near-real time by identifying relationships between unlabeled data elements rather than being trained on sequential labeled data.Sources:Activate analysis,Ars Technica,Dartmouth,Dataversity IBM,NVIDIA,QuartzGENERATIVE AI WILL LEAD TO THE KEY PARADIGM SHI

87、FTS OF THE FUTUREDeep LearningSEARCH&INFORMATION Search engines will shift towards conversational AI-driven interfaces that power online information searches and lead to a total transformation in discovery and search advertising for brandsCONTENT CREATION Text-to-text,-image,-video,and-audio tools w

88、ill be the lifeblood of developers,artists,marketers,and all creators,producing content at scale that is indistinguishable from that produced by humansENTERPRISE&THE WORKPLACE Text-to-code,administrative,process automation,and other AI tools will drive efficiencies,increase margins,make organization

89、s more lean,and fully automate workflows Machines are programmed to learn,make decisions,and improve using deep neural networks,powering a wave of new technology such as facial recognition,autonomous vehicles,and more1990s-2010sArtificial Intelligence/Machine LearningComputing shifts from input-outp

90、ut to being more intelligent and adaptive,as machines are programmed to perform complex tasksGenerative AILarge language and diffusion1 models enabled by transformers2 that are trained on large bodies of text/media showing emergent and human-like contentKEY EVENT:IBMs Deep Blue computer beats the wo

91、rld chess champion in 19971950s-1990sCurrentKEY EVENT:AlexNet Model wins ImageNet image classification challenge using convolutional neural network(CNN)architecture in 2012Generative AI will bring paradigm shifts across search,content creation,and enterprise use casesGENERATIVE AIGraphics processing

92、 units(GPUs)make previously-theoretical models computationally possibleGenerative adversarial networks(GANs)are developed,enabling deepfakes and face-swapping appsOpenAI is launched as a nonprofit AI research companyGoogle researchers introduce transformer models that become the architecture for mos

93、t of todays generative AILarge language(LLMs)and diffusion models using transformer architecture show emergent behavior,leading to the term“foundation models”JUNE 2009JUNE 2014DECEMBER 2015JUNE 2017AUGUST 2021Todays generative AI capabilities have been developing since at least 2009 29ACTIVATE TIMEL

94、INE OF MAJOR AI MILESTONESNote:Timeline is not sized to scale.Sources:Activate analysis,Analytics India Magazine,Dataversity,G2,Intelygenz,Medium,OctoML,Pitchbook,Reuters,TechCrunch,TELUS,Wired 29SEPTEMBER 2012MARCH 2015MAY 2016JANUARY 2021NOVEMBER 2022Convolutional neural networks(CNNs)emerge as be

95、st-in-class image classification models after AlexNet model wins ImageNet challengeDiffusion models are introduced,empowering breakthroughs in image generationGoogle develops its own chips,called Tensor Processing Units(TPUs),to accelerate the training of deep neural networksOpenAI releases text-to-

96、image diffusion model DALL-E,enabled by contrastive-language image pre-training(CLIP)After turning for-profit,OpenAI releases ChatGPT,the fastest-growing consumer platform in modern historyModel Research BreakthroughsBusiness Developments/Model ReleasesBEYONDIn Oct.2023,OpenAI will integrate DALL-E

97、3 with ChatGPT,creating the companys first multimodal consumer platform Google is expected to release Gemini by the end of 2023,to compete with OpenAIs GPT-4 Google is currently testing the integration of generative AI features into its Google Search engine Apple is rumored to make a significant AI

98、announcement in 2024GENERATIVE AIChatGPT has been adopted faster than any other platform in modern history(100M MAUs in two months)30Note:Analysis completed by evaluating the top 50 most visited websites(all time,as of Jan.27,2021,via Visual Capitalist)and top 50 apps with the most cumulative downlo

99、ads(all time,as of May 30,2023,via data.ai).List limited to companies that publicly announced surpassing 100M MAUs.Does not include announcements of total accounts,downloads,or(non-monthly)users.1.Analysis excludes Threads,as all Instagram users(2B+MAUs)were prompted to join Threads the day of its r

100、elease,July 6,2023,and Threads is not a“new platform,”as the platform is a clone of another app(X/Twitter).2.Refers to the international release of TikTok.3.CapCut is a video editing app launched by ByteDance in Apr.2020.4.Refers to the Google Maps mobile GPS navigation app,not the Google Maps websi

101、te.5.Qzone is a social networking website launched by Tencent in Apr.2005.Sources:Activate analysis,data.ai,Insider,Reuters,TechCrunch,Tech in Asia,Tencent,VentureBeat,Visual Capitalist 2 months 9 months1 year1 year,2 months1 year,11 months2 years,4 months2 years,6 months2 years,11 months2 years,11

102、months531 year,2 months42TIME TO REACH 100M MONTHLY ACTIVE USERS FROM LAUNCH BY PLATFORM1,GLOBAL,ALL TIME,YEARS AND MONTHSLAUNCH YEAR202220001320092005GENERATIVE AIAlthough ChatGPT web traffic has declined in recent months,traffic is increasingly shifting towards mobile;total m

103、onthly ChatGPT visits have doubled from April to August 2023 31Sources:Activate analysis,data.ai,OpenAI,SimilarWebTOTAL MONTHLY CHATGPT VISITS BY PLATFORM,U.S.,DEC.2022-AUG.2023,MILLIONS MONTHLY VISITSTotal Monthly Visits100M200M300M400MDec.Jan.Feb.Mar.Apr.MayJuneJulyAug.20222023App Visits(iOS and G

104、oogle Play)iOS mobile app launchWeb Visits(Mobile and Desktop)Total Monthly VisitsGENERATIVE AIAndroid mobile app launchAndroid mobile app launchChatGPT,Google Bard,and Bing Chat are seeing significant usage within a year of launch,while Adobe Firefly has exhibited fast growth in just a few months 3

105、2USAGE OF SELECT GENERATIVE AI TOOLS,U.S.,2023,%ADULTS AGED 18+/MONTHS1.“Current Users”are defined as those who use the generative AI tool at least once per month.2.Tool has created over 1 billion assets and supports over 100 languages.3.As of Aug.2023.Sources:Activate analysis,Activate 2023 Consume

106、r Technology&Media Research Study(n=4,023),Company sites,CrunchbaseMONTHS SINCE LAUNCH3CURRENT USERS11%2%3%4%5%6%02468012%13%2830323436Text-to-Text Text-to-ImageText-to-Video Text-to-Audio(12.0%)(3.9%)(4.9%)36+(1.5%)2GENERATIVE AI 33CONSUMER GENERATIVE AI TOOL USAGE,U.S.,2023,%MONTHLY GEN

107、ERATIVE AI USERS11.“Monthly generative AI users”are defined as adults aged 18+who use at least one generative AI tool at least once per month.2.Includes generative AI tools primarily known in a multipurpose way.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)

108、,Company sitesGeneral-Use Tools2USE CASES%MONTHLY GENERATIVE AI USERS1EXAMPLESEXAMPLE TOOLSWriting ImprovementProofread material,edit an emailEntertainmentCreate a game,personalize content on social appsContent CreationGenerate an image,create an avatar in a video gameConcept ExplanationExplain comp

109、lex science concepts with a personal tutorSelf-Help/ImprovementProvide motivation,act as a soundboard for personal issuesFind RecommendationsSummarize customer reviews,suggest vacation destinations Personal AssistantSchedule meetings,make a budgetTranslatingTranslate languagesDigital DesignCreate a

110、web interface,improve a PowerPointCareer AdvancementCreate interview questions,generate a resumeCodingGenerate code from plain language,correct code errorsTranscribingTranscribe calls,add subtitles to videosOther10%14%16%19%19%20%20%21%22%23%25%29%30%Consumers are using generative AI tools to suppor

111、t a broad range of personal and professional use cases todayGENERATIVE AIOver half of people who are aware of generative AI tools today intend to use them in the future 341.“Aware of generative AI”is defined as knowing what generative AI tools are.2.“Current users”are defined as those who use at lea

112、st one generative AI tool at least once per month.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)GENERATIVE AI AWARENESS1,U.S.,2023,%ADULTS AGED 18+GENERATIVE AI INTENT TO USE BY CURRENT USAGE,U.S.,2023,%ADULTS AWARE OF GENERATIVE AI158%42%Not Aware of Gener

113、ative AI1Aware of Generative AI1Intend to use in next 12 monthsDo not intend to use in next 12 monthsCurrent Users2 37%Non-Users2 18%Current Users2 2%Non-Users2 43%55%of those aware of generative AI1 intend to use it in the next 12 monthsGENERATIVE AIPeople are most interested in using text-to-text

114、and text-to-image generative AI tools 35INTENT TO USE GENERATIVE AI BY AI TOOL TYPE,U.S.,2023,%ADULTS AGED 18+WHO INTEND TO USE GENERATIVE AI11.“Intending to use generative AI”is defined as indicating intent to use any generative AI tool in the next 12 months.Sources:Activate analysis,Activate 2023

115、Consumer Technology&Media Research Study(n=4,023)TEXT TO TEXT (i.e.chatbots)TEXT TO IMAGETEXT TO AUDIOTEXT TO VIDEO33%35%47%68%GENERATIVE AI 36FUNDAMENTAL BUILDING BLOCKS FOR GENERATIVE AISources:Activate analysis,Analytics Insight,Daffodil,H20 AI,TechCrunch,Tech Monitor,UX Collective,Web Strategist

116、MODELSDATAWHY:Models are the algorithms that receive inputs,process them,identify patterns,and ultimately create output WHO WILL BENEFIT:Historically,major technology companies and well-funded startups,but currently an area of disruptionWHY:Data with sufficient breadth and depth is critical to train

117、 AI models,with more/higher-quality data resulting in better output WHO WILL BENEFIT:Training infrastructure providers,major technology companies with vast customer data,and industry-specific players with access to highly-tailored data(e.g.healthcare)COMPUTE&PHYSICAL INFRASTRUCTUREINTERFACES&SOFTWAR

118、E TOOLSWHY:AI models can be very computationally expensive,even when using optimized hardware such as GPUs or ASICs WHO WILL BENEFIT:Chip companies,foundries,cloud providers,and companies building AI supercomputersWHY:Strong interfaces are needed for adoption,especially for areas with high complexit

119、y of rules(e.g.law,healthcare),given AIs stochastic nature;tools are also needed for databases,fine-tuning models,and data generation/labeling(among others in MLOps)WHO WILL BENEFIT:Safety-focused AI companies(i.e.those with a focus on content moderation),advanced industry-specific interfaces,data i

120、nfrastructure tools,and tailored use-case-specific interfaces ExamplesExamplesExamplesExamplesGenerative AIThere are four fundamental building blocks to consider within the greater ecosystem of generative AI,with different companies expected to benefit from the importance of each:models,data,compute

121、,and interfaces&software toolsGENERATIVE AIONGOING DISRUPTION:OPEN-SOURCE MODELS LEADING TO COMMODITIZATIONThe generative AI playing field will not be limited to the major technology players,particularly with the rise of open-source AI models opening the door for new,smaller entrants 37STEPS TOWARDS

122、 COMMODITIZATIONPre-trained models have become publicly available:Meta models(LLaMA,LLaMA 2,and Code LLaMA),Falcon LLM,and StableStudio are prominent examples As of August 2023,Hugging Face has over 300,000 available open-source modelsOpen-source models become increasingly high-quality and low-cost:

123、Training for LLMs such as GPT-4 are estimated at upwards of$100M1 Training costs for recent open-source models can be in the hundreds (e.g.$600 for Alpaca)Open-source communities rapidly innovate and improve models:Researchers in the open-source community are improving models faster(by some accounts

124、)than many major technology companies Models are already showcasing similar results with much lower costs(e.g.Vicuna-13B achieving 90%of GPT-4 quality for$300 in training)STRATEGIC IMPLICATIONSIndependence from Major Technology Players Companies do not need to go to major technology players such as

125、Meta,Google,or Microsoft for paid access to high-quality generative modelsMore Accessibility for Custom Models Companies can create more custom high-quality generative AI models for significantly lower costsPotential Commoditization of Models as Base Layer Generative AI models may follow a path simi

126、lar to technologies like Linux,where paid applications are built off of a free and open-source base1.In Apr.2023,OpenAI CEO Sam Altman claimed the full cost of training GPT-4 was over$100M.Sources:Activate Analysis,Databricks,Financial Times,Forbes,Hugging Face,The Information,Large Model Systems Or

127、ganization,MIT Technology Review,MosaicML,New Atlas,The Register,TechCrunch,WiredGENERATIVE AI 381.“Using generative AI as a first stop for online search”is defined as typically starting search at a generative AI tool when looking for information online.Includes normal search with a generative AI su

128、mmary at the top of a search results page.2.Reflects Aug.2023.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023),U.S.Census Bureau 3813M90M202322027EUSAGE OF GENERATIVE AI AS A FIRST STOP FOR ONLINE SEARCH1,U.S.,20232 VS.2027E,MILLIONS ADULTS AGED 18+ACTIVATE F

129、ORECASTToday,over 10M U.S.adults begin their web search with generative AI;this will grow to 90M by 2027,promising to reshape the largest technology businessesGENERATIVE AIBrands across a broad set of industries have already integrated generative AI into their consumer offerings 39ACTIVATE TIMELINE

130、OF AI CONSUMER BRAND OFFERINGS,U.S.,JAN.2023-SEPT.20231Note:Order of AI offerings within month and size of timeline lines are for illustrative purposes,not indicative of exact announcement dates or importance.Timeline is not exhaustive.1.As of Sept.6,2023.Sources:Activate analysis,ABC News,Businessw

131、ire,CNBC,CNN,Company press releases,Company sites,Hospitality Net,The Motley Fool,NASDAQ,PRNewswire,Retail Brew,Retail Dive,Skift,TechCrunch,TVTech,VentureBeat,The Verge,The Wall Street Journal,The Washington Post,ZDNetChatbot Offering New/Upgraded Search Capabilities New/Upgraded Other Product or F

132、eature(e.g.AI Audio Narration)(e.g.Automated Vehicle Assessment)(e.g.AI DJ)(e.g.Text-Based AI)(e.g.AI Language Roleplay)(e.g.AI-Powered Writing)(e.g.MS 365 Copilot)(e.g.“Create Real Magic”Contest)(e.g.ChatGPT Plug-in)(e.g.Shopify Commerce Components Integration)(e.g.Brand Hub)(e.g.“Shopify Magic”)(e

133、.g.Petey AI With Apple Music)(e.g.Personal Recommendations/GPT4 Integration)(e.g.“MusicLM”)(e.g.AI WordPress Assistant)(e.g.AI Text-Based Reply)(e.g.“Project Tailwind”)(e.g.AI-generated quizzes)(e.g.AI Trip Planner)(e.g.“Dropbox AI”)CHATBOT OFFERINGSEARCH CAPABILITIESOTHER PRODUCT/FEATUREInteractive

134、 chat tool for individual use across search and social platforms,leveraged for enhanced functionality Examples include Snapchats“My AI”and Googles“Bard”Ability to leverage generative AI for enhanced and interactive search functions Examples include Instacarts“Ask Instacart”recipe generator and Yelps

135、“Surprise Me”search recommendation toolSupplementing current technology with a new or improved AI-powered offering Examples include Spotifys new AI DJ and Apples AI-enhanced book narration(e.g.“Dropbox Dash”)(e.g.“Decorify”)(e.g.Customer Feedback Summaries)(e.g.Automatic Description Generation)(e.g.

136、“Generative AI Innovation Center”)(e.g.“HubSpot AI”)(e.g.“Cue”AI Assistant)(e.g.Online Store Product Descriptions)(e.g.“Freddy Self Service”)(e.g.“Tongyi Wanxiang”)(e.g.AI Itinerary Generator)(e.g.“Generative AI Product Accelerator)(e.g.“Y3000”Flavor)(e.g.“ContractFlow”)(e.g.“Rapid Recap”)(e.g.Googl

137、e Cloud/ELC Collaboration)(e.g.AI Checkout Feature)(e.g.“Magic Notes”)JAN.MAR.APR.MAYJULYAUG.FEB.SEPT.JUNEGENERATIVE AICREATIVE APPLICATIONEXAMPLE USE CASESEXAMPLE TOOLSARTCreating detailed digital sketchesCreating film posters or album cover artworkAUDIOCreating background music for a songVoice clo

138、ning enabling digital voice generationVoices optimized to read different genres of audiobooksINSPIRATIONGenerating an outline for online coursesGenerating a list of ideas and prompts for content creationGenerating images as a starting point for artistic creationsVIDEORemoving filler wordsGenerating

139、visuals to complement audio filesDubbing a video in different languagesClipping long-form videos into shorter videosOver 60%of creators already use generative AI today 401.“Content creators”are defined as those who edit and share photos or videos online that generated income.Sources:Activate analysi

140、s,Company sites,Forbes,The Information,Lightricks/YouGov 2023 Creator Economy Report (n=1,000+),Mashable,Stanford Institute for Human-Centered Artificial IntelligenceGENERATIVE AI AWARENESS,USAGE,AND RATIONALE FOR USAGE,U.S.,MAR.2023,%CONTENT CREATORS1GENERATIVE AI USE CASES54%84%84%86%Generative AI

141、 Familiarity Among Content Creators1Generative AI Usage Among Content Creators1Helps the Creative ProcessSaves TimeSaves MoneyRequest From a BrandCreator Rationale%Content Creators1 Who Have Used Generative AI362%92%GENERATIVE AIITERATIVE CREATION JOURNEY VALUE CHAIN 41Source:Activate analysisThe ca

142、pabilities of generative AI will revolutionize the creation journey value chain across media and entertainmentMEDIA TYPEEXAMPLE USE CASESGENERATIVE AIIDEATIONINITIAL CREATIONEDITING/REFINEMENTPERSONALIZATIONPROMOTION/MARKETINGAGGREGATION OF CONSUMER BEHAVIOR AND FEEDBACKVIDEOList of video ideas Scri

143、pt writing Concept artInitial video creationSegmenting videos into clips Removing filler words Aiding in visual effect productionSubbing/dubbing video in different languagesFilm posters Movie trailersSummarized feedback from movie reviewsMUSICList of song ideasSong lyrics/beats Voice cloning Visuals

144、 to complement audioAltering style to match a different genre/artistLanguage translationAlbum cover artSynthesized commentary from sources such as YouTube commentsOTHER AUDIO(e.g.podcasts,audiobooks)List of podcast ideas Podcast script writingAudio based on written material(e.g.artificial voice)Remo

145、ving filler words Cutting out dead audio/upscaling audio qualityVoices optimized to read different genres of audiobooks Automated transcripts of podcast episodes Language translationAudio ad creationActionable feedback from app store reviewsGAMINGConcept art for in-game environments and charactersNP

146、C dialog/story development Virtual world creation AI-driven questsChat moderation Copilot for coding Automated bug detectionAvatar and object creation/customizationGame trailer creationAnalytics enhanced with AI to better assess player behavior and preferencesART/IMAGESNearly instant initial image c

147、reationText-to-image detailed sketches/imagesExisting image modifications(e.g.recolor)Background fill Extend imagesCustom image modifications based on text promptsOrganize/manage assets for better marketingInput such as conversion factors on marketing to inform popular contentTEXT (e.g.articles)List

148、 of article,quiz,or other text ideasInitial article writingAltering text tone/style Outline/summary of textLanguage translationHeadlines optimized for trafficFeedback from social media/article commentsGenerative AI is also expanding into the workplace,especially among younger and higher-income emplo

149、yees;16%use generative AI in their workplace today,with 21%expecting to use it in the next 12 months 42ROBOTO BOLD 16PT,ROBOTO REGULAR 16 PTAWARENESS1 VS.USAGE VS.EXPECTED USAGE2 OF GENERATIVE AI IN THE WORKPLACE,U.S.,2023,%EMPLOYED ADULTS3AWARENESS OF GENERATIVE AI1USAGE OF GENERATIVE AI IN THE WOR

150、KPLACE IN THE LAST 12 MONTHSEXPECTED USAGE2 OF GENERATIVE AI IN THE WORKPLACE IN THE NEXT 12 MONTHSEmployed adults3 with annual household income of$100K+are as likely to have used generative AI at work than those with annual household income$100KEmployed adults3 aged 18-44 are as likely to have used

151、 generative AI at work than those aged 45+16%84%48%52%Have not used at workAwareNot aware21%79%Do not expect to use at work2.5x2.4xNote:Images created using Midjourney.1.“Aware of generative AI”is defined as knowing what generative AI tools are.2.“Expected usage”is defined as those expecting generat

152、ive AI to be a part of their job in the next 12 months.3.“Employed adults”are defined as adults aged 18+that are employed full-time,employed part-time,self-employed,or in the military.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023),MidjourneyGENERATIVE AIExp

153、ect to use at workHave used at work 43CURRENT USAGE VS.EXPECTED USAGE1 OF GENERATIVE AI IN THE WORKPLACE BY SELECT JOB ROLE2,U.S.,2023,%EMPLOYED ADULTS3 BY JOB ROLE2PERCENTAGE POINT DIFFERENCE4AVERAGE WORKPLACE USAGE5:16%JOB ROLE2 INFORMATION TECHNOLOGYENGINEERINGACCOUNTING/FINANCEARTS AND DESIGNMAR

154、KETINGBUSINESS DEVELOPMENTADMINISTRATIVE /HUMAN RESOURCESEDUCATIONTOP 2 USE CASES#1DATA ANALYSISDATA ANALYSISDATA ANALYSISCONTENT CREATIONCONTENT CREATIONCONCEPT EXPLANATIONTIED ACROSS MULTIPLE USE CASESCONCEPT EXPLANATION#2SKILL DEVELOPMENTSKILL DEVELOPMENTPERSONALIZATIONDIGITAL DESIGNTIED ACROSS M

155、ULTIPLE USE CASESPROCESS OPTIMIZATIONADMINISTRATIVE TASKS6pp2pp7pp12pp16pp10pp7pp2ppLargest Expected Growth16%24%31%39%38%34%36%48%14%17%20%23%26%27%34%42%Current UsageExpected Usage11.“Expected usage”is defined as those expecting generative AI to be a part of their job in the next 12 months.2.Inclu

156、ded job roles based on those with highest self-reported generative AI usage.3.“Employed adults”are defined as adults aged 18+that are employed full-time,employed part-time,self-employed,or in the military.4.Figures do not sum due to rounding.5.Represents average workplace usage of generative AI acro

157、ss all job roles.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)Knowledge-based and creative functions,such as information technology and marketing,are seeing the greatest generative AI use along with high expected future growthGENERATIVE AIDATA COLLECTIONPR

158、OCESS OPTIMIZATIONCODINGExtracts specific visual and text data from websites,converting it into spreadsheet form for the userCreates simple,user-friendly,automated software documentation across a number of programming languagesGenerates code for programmers to build,test,and deploy directly from the

159、ir browserCreates data visualizations,builds dashboards,and presents data for agencies,marketers,eCommerce companies,and moreAllows non-developers to collect,parse,act on,and share raw data within their workplace with a no-code toolMonitors and analyzes brand conversations synthesizing data across s

160、ocial,broadcast,digital,and traditional media channelsAids in the cleaning,building,and analyzing of datasets by plugging into Google Sheets and other toolsGenerative AI work use cases:Data,Process,and Coding 44EXAMPLE AI WORK USE CASES AND EXAMPLE COMPANIES AND TOOLSSources:Activate analysis,Compan

161、y sitesDATA MANAGEMENTDATA ANALYTICSGENERATIVE AIADMINISTRATIVE TASKSCUSTOMER SUPPORTSupports scheduling,collaboration,and productivity needs for enterprise teams through AI“Chief of Staff”conversational assistantPrevents,detects,responds to,and predicts email and email-like fraud attacks through be

162、havioral AI engineProvides customer-facing voice assistants and generates relevant insights and potential operational improvements to internal customer support teamsCreates company-specific learning plans to upskill employees in software,financial services,and other industriesDetects and mitigates f

163、raud by monitoring the marketplace,detecting the start of fraud,and generating rules to stop it through“Project RADAR”Offers customer support chatbots while powering customer support teams through assisted content generation,training suggestions,and generative repliesAccelerates employee acquisition

164、 of internal data and knowledge with chatbots trained on company dataCONCEPT EXPLANATIONProvides explanations on complex topics and serves as a virtual tutor with generalist chatbotGenerative AI work use cases:Support/Training and Moderation/Fraud Detection 45EXAMPLE AI WORK USE CASES AND EXAMPLE CO

165、MPANIES AND TOOLSSources:Activate analysis,Company sitesSKILL DEVELOPMENTMODERATION/FRAUD DETECTIONGENERATIVE AIGenerative AI work use cases:Creation and Design 46EXAMPLE AI WORK USE CASES AND EXAMPLE COMPANIES AND TOOLSSources:Activate analysis,Company sitesPRODUCT INNOVATIONDIGITAL DESIGNPERSONALI

166、ZATIONHelps companies predict and understand protein structure with Target and Lead Identification SuiteGenerates wireframes,mock-ups,and prototypes for UI designersCreates personalized content emails and other content using Salesforce CRM data through Einstein GPT toolTranslates 2D sketches into 3D

167、 photorealistic renders for prototypers or product designersCreates presentations with premade templates and themes with a collaborative AI assistant using OpenAIs GPT-4 and DALL-E 2Automates entire outreach process for email marketers with personalized messaging,icebreakers,and contextual pitchesAl

168、lows advertisers to produce high-quality commercial creative including images,videos,and 3D renders through partnership with NVIDIAGenerates and tests copywriting for written advertisements using proprietary large language modelsGenerates game assets such as characters,environments,and items for vid

169、eo game designers in addition to other forms of 3D artCONTENT CREATIONGENERATIVE AI 47The AI ecosystem has exploded in the past year;many of the current companies will not last as the space continues to grow more competitiveINTERFACES&SOFTWARE TOOLSDATA(TOOLS&SOFTWARE)DATAMODELS&MODEL DEVELOPMENTCOM

170、PUTE&PHYSICAL INFRASTRUCTURESOFTWARE INFRASTRUCTURE&MODELSHARDWAREAI CybersecuritySynthetic DataData ManagementData LabelingModels2Training/Development/DeploymentCloud Service Providers(CSP)Chip Designers&ManufacturersIncludes specific open source models and companies that have developed and release

171、d at least one open source modelMUSICText-to-MusicMisc.Music CODING/SOFTWARE DEVELOPMENTAutomated DocumentationWeb/App BuildersCoding CopilotsFalcon LLMSPEECHSpeech/Audio TranscriptionText-to-SpeechSALES&MARKETINGCopywritingSEOGraphic DesignSalesUI/UX DesignEmail/SMS AutomationWRITINGWriting Assista

172、nts/AugmentationTranslationGAMINGGaming ExperiencesCharacter/Avatar Development3D World/Asset DevelopmentOTHER B2B/ENTERPRISE INTERFACESOther B2BProptechHealthcareCybersecurityLegalCustomer SupportIndustrialADMINISTRATIVE/HRDocument Processing/SummaryRecruitingExecutive AssistantsCollaborationVIDEOM

173、isc.VideoText-to-VideoVideo EditingACTIVATE GENERATIVE ARTIFICIAL INTELLIGENCE ECOSYSTEMGENERALIST CHATBOTS1My AIIMAGEMisc.ImageImage EditingText-to-Image1.Includes chatbots primarily known in a multipurpose way(e.g.for search,companionship).2.Includes models and companies/divisions developing model

174、s.Sources:Activate analysisGENERATIVE AIMajor technology companies are building capabilities across the AI tech stack but are not currently fully leveraging their consumer data 48=Open Source=Closed Source,Available Through API=Closed Source,Not Available Through API=Not Yet DisclosedNote:Not exhaus

175、tive.As of Oct.10,2023.1.Apples system runs on Google Cloud,which Apple uses to power cloud services alongside its own infrastructure and AWS.2.As part of its partnership with OpenAI,Microsoft is OpenAIs exclusive cloud provider.The increased workloads on Azure represent a significant revenue opport

176、unity.3.A custom-designed chip to specifically address AI problems.4.“Tensor Processing Units(TPUs)”are Googles custom-developed application-specific chips used to accelerate machine learning workloads.5.Apple develops its own“systems-on-chips”for its devices(e.g.M2,A17),but no AI-dedicated chips.6.

177、An interface that allows developers to build and deploy machine learning models more quickly and easily.7.It is reported that Apple is testing a new chatbot;different entities refer to it as Apple GPT.8.Amazon announced generative AI features for its Alexa products powered by a new multimodal Alexa

178、LLM in September 2023 that have not yet been released.9.Benefits include access to ChatGPT during peak times,faster response times,and priority access to new improvements.10.Only refers to Alexa LLM.Users can opt in to having their voice data and conversations used for training the Alexa LLM.Not rev

179、ealed for Titan models.11.WebText2 is a dataset created using links from Reddit posts that have at least three upvotes.Sources:Activate analysis,Business Insider,Company sites,Financial Times,Reuters,The VergeNETFLIXNVIDIAOPEN AIGENERATIVE AI INFRASTRUCTURE AND HARDWARECLOUD SERVICE PROVIDERGoogle C

180、loud PlatformAmazon Web ServicesN/A1Microsoft Azure2Nvidia DGX CloudAI CHIP3AWS Trainium AWS InferentiaN/A5AthenaH100 Tensor Core GPUCORE SOFTWARE&GENERATIVE AI MODELSML FRAMEWORK/PLATFORM6Microsoft Cognitive ToolkitNvidia DIGITSMULTIMODAL MODELGeminiLeveraging models from:Apple GPT6Leveraging model

181、s from:TEXT-TO-TEXT MODELMegatron-Turing NLGTEXT-TO-IMAGE MODELLeveraging models from:TEXT-TO-AUDIO MODELTEXT-TO-VIDEO MODELGENERATIVE AI PRODUCT OFFERINGSAPP STOREGenerative AI apps available in Play StoreGenerative AI apps available in App StoreThe Microsoft Store introduced its AI hub in the apps

182、 left sidebarReportedly planning to offer a generative AI app storeCONSUMER OFFERINGSBard BERT PaLM 2Bing AIChatGPT ChatGPT Plus7ENTERPRISE OFFERINGSDATA INPUTSDATA USED TO TRAIN GENERATIVE AI MODELSCommon Crawl,Wiki,code,public forum dialogMultilingual speech data,opted in Alexa user-device convers

183、ations10Reportedly Common Crawl,web articles,and booksPublicly-shared Facebook and Instagram posts,Common Crawl,books,WikiNot reportedCommon Crawl,WebText211,books,WikiChatGPT user interactions,Common Crawl,WebText211,books,and WikiEXAMPLE AVAILABLE DATA (BUT NOT CURRENTLY USED TO TRAIN GENERATIVE A

184、I MODELS)YouTube videos Search and browsing behavior Maps and geolocation App photos and messages Online transactionseCommerce transactions Prime Video and Fire TV viewingDevice messages and photos Search App activity Siri device audioApp transactions Geolocation Facial recognition Private Facebook

185、and Instagram posts and messagesSearch Geolocation Apps activityGeolocation Network activity2=Announced,not yet released7(e.g.TPU v5)49Enterprise3LLM8GENERATIVE AICATEGORYPRIMARY FOCUSEXAMPLESBROAD USE CASESDevelop LLM platforms and search engines for generalized use casesB2B APPLICATIONSBuild enhan

186、ced systems leveraging AI to improve business productivity and enterprise functionsSOFTWARE INFRASTRUCTURELeverage machine learning/natural language models to develop AI tools across verticalsCONSUMER APPLICATIONSGenerate AI services and platforms to assist end customers/users HARDWARECreate advance

187、d computing infrastructure and devices to power scaled intelligence models and accelerate processing speedsGENERATIVE AI TOTAL181%YoY growth(vs.-42%YoY growth for VC investments overall)As of August 2023,generative AI-related ventures saw nearly 200%growth YoY,with broad use case platforms,software

188、infrastructure,and B2B applications making up over 95%of investments 49INVESTMENTS IN GENERATIVE AI BY CATEGORY,GLOBAL,SEPT.2021-AUG.2023,BILLIONS USD$19.8B$0.2B$0.5B$3.0B$3.3B$12.9B$7.1B$0.5B$0.7B$1.4B$3.3B$1.2BSept.2021-Aug.2022Sept.2022-Aug.2023Note:Amazons acquisition of iRobot in Aug.2022 for a

189、 value of$1.7B is excluded from this deal set.Microsofts acquisition of Nuance Communications Inc.in Mar.2022 for a value of$19.7B is excluded from this deal set to adjust for extreme outliers.Sources:Activate analysis,PitchbookGENERATIVE AICompanies will not need to build AI in-house to leverage th

190、e benefits;major AI companies are building plug-ins and integrations to bring AI to a broader ecosystem 501.Initial reporting suggests that plug-ins have thus far not proved popular with consumers.Sources:Activate analysis,Company sites,Comscore,The Information,OpenAI,Reuters,Similarweb,Statista,Tec

191、hCrunchTHIS PATHWAY IS DRIVING INCREASED EXPERIMENTATION/INVESTMENT IN GENERATIVE AI AMONG MEDIA COMPANIESTubi announced in September 2023 that it was partnering with Rabbit AI to offer a chatbot-based search function to allow users to go beyond simple search keywords to discover video content on it

192、s platformGoogle announced an initiative in May 2023 to leverage generative AI(specifically their PaLM LLM)to automate advertising services and suggest videos for YouTube creators to developComcast NBCUniversal launched the LIFT Labs:Generative AI Accelerator in May 2023 to partner with/develop star

193、tups working across the generative AI ecosystemCONSIDERATION 1:SOFTWARE LICENSING/DATA AGREEMENTSCONSIDERATION 2:PLUG-INS/INTEGRATIONSCONSIDERATION 3:SUBSCRIPTIONSThe approach on both data and underlying models are core to any AI partnership:Models offered by AI companies can be either closed source

194、(e.g.GPT-4)or open source(e.g.LLaMA)Data from licensee can be used to customize/supplement AI models Data from end users can continually fine-tune training Data control policies are critical to ensure sensitive data(e.g.PII)is protected and used appropriatelyThere are two primary paths to integrate

195、AI into existing products:Integrate AI functionalities into a third-party source(e.g.“Ask Instacart”feature within app)Embed product into existing AI interface such as ChatGPT as a plug-in1(e.g.Expedia plug-in expanding ChatGPT trip planning capabilities)The pricing structure for AI follows an“aaS”s

196、ubscription model;companies can integrate AI into subscriptions in two ways:Restrict select AI tool features (e.g.priority access to new integrations,faster response times)to higher tiers(e.g.ChatGPT Plus)Embed AI as a feature/functionality within existing platforms(e.g.Microsoft Designer applicatio

197、n)PATH FOR COMPANIES CONSIDERING AI INTEGRATIONS WITH EXISTING APPLICATIONS AND SERVICESGENERATIVE AIDespite the promise,there are barriers to realizing the full market potential of generative AI 51Sources:Activate analysis,Axios,BBN Times,Company sites,eMarketer,Financial Times,Insider Intelligence

198、,LeewayHertz,OpenAI,TechCrunch,Telus International,Time,The Verge,The Wall Street JournalGovernments may enact policies to exercise significant oversight and regulatory control on AIPOTENTIAL REGULATIONIt is unclear if AI-generated content can be protected under copyright law,potentially slowing ado

199、ptionINABILITY TO COPYRIGHT OUTPUTMajor language models may eventually run out of optimal training dataLACK OF TRAINING DATAPlatforms do not always reject prompts that can lead to individual or societal harmSAFETY CONCERNSAI platforms owning company data potentially allows them to become a competiti

200、ve threatCOMPETITIVE CONSIDERATIONSSensitive and private data(e.g.PII)used in training can sometimes leak into outputDATA SECURITY/PRIVACYHuman-made input datasets used for training generally contain difficult-to-combat biasesMODEL BIASThere are several legal challenges to generative AI due to poten

201、tial IP violations(e.g.training data)IP CHALLENGESGiven the stochastic nature of generative AI,it is prone to factual errorsINACCURACIES/MISINFORMATIONGENERATIVE AI 52CONTENTSPAGEConsumer Attention5Super Users10Generative AI26eCommerce52B2B and Enterprise Software64Gaming84Metaverse102Spatial Comput

202、ing 118Social Video129Video135Music and Podcasts160Sports Betting177Live Entertainment and Experiences184Technology and Media Revenues195Global eCommerce will reach nearly$10T by 2027,growing more than twice as fast as physical retail 531.Excludes travel and event tickets,food and drink services,and

203、 vice goods and activities.2.Figures do not sum due to rounding.Sources:Activate analysis,eMarketer,Research and Markets,StatistaRETAIL SALES1 BY CHANNEL2,GLOBAL,2023E VS.2027E,TRILLIONS USDeCommercePhysical RetailACTIVATE FORECAST2023E-2027E CAGR:28%eCommerce Share of Total Retail Sales12023E2027E$

204、33.8T$27.4T$9.6T$3.1T$6.5T$3.2T$24.2T$21.0T4%10%5%24%ECOMMERCEGlobal eCommerce merchandise volume is heavily concentrated among a small group of players 54SHARE OF ONLINE GROSS MERCHANDISE VOLUME(GMV)1 BY COMPANY,GLOBAL,2022,%TOTAL ONLINE GMV11.Excludes travel and event tickets,food and drink servic

205、es,and vice goods and activities.2.Each company accounts for less than 1%of the total online GMV.3.Includes all of online GMV not covered by the top 20 eCommerce companies.Sources:Activate analysis,Company filings,Company press releases,Company sites,Digital Commerce 360,eMarketer,The Information,St

206、atista,U.S.Internal Revenue Service35%4%8%9%12%23%Rest of Web3$5.8T 2022 Total Global Online GMV1Pinduoduo1%1%1%4%261%of total global online GMV1 is generated by the top 10 eCommerce companies1%Douyin1%ECOMMERCEeCommerce is a marketplace business;third-party sellers generate over 80%of online sales

207、for the top 10 eCommerce players 551.Excludes travel and event tickets,food and drink services,and vice goods and activities.Sources:Activate analysis,Company filings,Company press releases,Company sites,Digital Commerce 360,eMarketer,The Information,Research and Markets,Statista,U.S.Internal Revenu

208、e ServiceSHARE OF ONLINE GMV1 OF TOP 10 ECOMMERCE COMPANIES BY SALES TYPE,GLOBAL,2022,%TOTAL ONLINE GMV1 OF TOP 10 ECOMMERCE COMPANIESSHARE OF ONLINE GMV1 BY SALES TYPE FOR TOP 10 ECOMMERCE COMPANIES,GLOBAL,2022,%TOTAL ONLINE GMV1 BY COMPANY$3.5T 2022 Total Global Online GMV1 of Top 10 eCommerce Com

209、panies81%Marketplace (Third-Party Sales)BUSINESS MODELECOMMERCE COMPANYMARKETPLACE LAUNCH YEAR2022 GLOBAL ONLINE GMV1 MARKETPLACE VS.DIRECT RETAILMARKETPLACE ONLY2003200151997HYBRID9DIRECT RETAIL ONLYN/APinduoduo9 of the top 10 global eCommerce players operate a marketplace19%D

210、irect Retail (First-Party Sales)63%37%52%48%11%89%Douyin100%100%100%100%100%100%100%ECOMMERCE 561.As of Oct.2023.Sources:Activate analysis,Company filings,Company press releases,Company sitesIn the U.S.,a growing number of major eCommerce companies beyond Amazon and Walmart are launching their own m

211、arketplaces to enhance their customer value proposition and retail media potentialPOWERED BYSTRATEGIC DRIVERS FOR LAUNCHING A MARKETPLACEECOMMERCE COMPANYMARKETPLACELAUNCH YEAR2002220232023EXAMPLE MARKETPLACE LAUNCHES BY MAJOR ECOMMERCE COMPANIES IN THE U.S.OVER THE LAST 5 YEARS1Enable or

212、 accelerate the retail media opportunityCreate new revenue streams through marketplace seller feesIncrease product assortment in existing categories to increase customer engagement and spendExpand into adjacent categories to build a more holistic offering and acquire new customersTest customer deman

213、d for new products with limited inventory risk and minimal costsLocalize supply chains for international retailers to improve delivery speedKrogers marketplace products are seamlessly integrated into its website/app and can be purchased with non-marketplace productsECOMMERCEU.S.retail media revenue

214、is expected to more than double over the next four years,reaching nearly$100B by 2027;we expect most major retail companies will have their own retail media network 57RETAIL MEDIA NETWORK ADVERTISING REVENUE1 BY COMPANY2,U.S.,2019 VS.2023E VS.2027E,BILLIONS USD20192023E2027E$97B$44B$13B$68B$33B$10BR

215、est of Market2019-2023E CAGR:$3B$8B$0.5B$3B$10BACTIVATE FORECAST$19B1.Includes digital advertising that appears on a retailers on-premise signage,either in-store or in the immediately adjacent exterior(e.g.parking lot),appears on websites or apps that are primarily engaged in eCommerce,or is bought

216、through a retailers media network or demand-side platform.2.Figures do not sum due to rounding.Sources:Activate analysis,Company filings,Company press releases,Company sites,eMarketer,Forrester,GroupMEXAMPLE MAJOR RETAIL COMPANIES THAT HAVE LAUNCHED A RETAIL MEDIA NETWORK 60%32%35%2023E-2027E CAGR:2

217、0%35%24%22%34%ECOMMERCETodays consumers are multi-channel shoppers,using a range of online channels for shopping inspiration and research 58ROBOTO BOLD 16PT,ROBOTO REGULAR 16 PT1.“Online shoppers”are defined as adults aged 18+who shopped online at least once in the last 12 months.Shopping includes b

218、rowsing and making purchases.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)When looking online for shopping inspiration and ideas,I typically useWEBSITES/APPSWhen doing online research on the items that I am considering purchasing,I typically useSearch engi

219、ne (e.g.Google,Bing,Yahoo)Social media platform (e.g.TikTok,Facebook,Instagram)Specific brands online store (e.g.Apple,Nike,Fenty Beauty)Membership warehouse club (e.g.Costco,Sams Club,BJs)Other online store that sells multiple brands (e.g.Best Buy,Foot Locker,Ulta)Other online marketplace that offe

220、rs items from multiple sellers(e.g.OfferUp,Etsy,Poshmark,Wayfair)Online delivery service (e.g.Instacart,Shipt,FreshDirect,DoorDash)4%6%9%11%15%13%19%27%50%24%41%67%WEBSITES/APPS TYPICALLY USED FOR SHOPPING INSPIRATION AND RESEARCH,U.S.,2023,%ONLINE SHOPPERS181%52%34%31%25%18%16%16%13%10%9%31%ECOMMER

221、CETikTok is disrupting online shopping,quickly emerging as a top inspiration and research source among younger consumers 59TOP WEBSITES/APPS TYPICALLY USED FOR SHOPPING INSPIRATION AND RESEARCH BY AGE GROUP,U.S.,2023,%ONLINE SHOPPERS BY AGE GROUP1Capitalizing on its popularity as a source for online

222、 shopping inspiration and research,TikTok officially launched its TikTok Shop marketplace nationwide in September 2023Top 5 websites/apps typically used for shopping inspiration and ideasTop 5 websites/apps typically used for doing research on items being considered for purchaseAged 18-34Aged 35-54A

223、ged 55+Aged 18-34Aged 35-54Aged 55+178%83%82%62%70%69%259%56%44%45%Search engine (e.g.Google,Bing,Yahoo)54%Search engine (e.g.Google,Bing,Yahoo)51%348%39%Search engine (e.g.Google,Bing,Yahoo)30%Search engine (e.g.Google,Bing,Yahoo)44%44%34%4Search engine (e.g.Google,Bing,Yahoo)28%Search engine (e.g.

224、Google,Bing,Yahoo)34%28%33%28%20%524%26%20%22%19%14%1.“Online shoppers”are defined as adults aged 18+who shopped online at least once in the last 12 months.Shopping includes browsing and making purchases.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023),Compan

225、y press releasesECOMMERCE 60CHALLENGES OF MULTI-CHANNEL MANAGEMENTBRANDING AND VALUE PROPS Maintaining the same brand identity across online channels DEMAND&SUPPLY PLANNING Optimizing assortment and inventory allocation across online channels PRICING Ensuring consistent prices and promotions across

226、online channels EXPERIENCE Coordinating across online channels to deliver a uniform customer experience DATA ANALYTICS Unifying customer data and insights across online channelsWEBSITES/APPS PURCHASED FROM IN THE LAST 12 MONTHS,U.S.,2023,%ONLINE PURCHASERS110%11%15%15%19%26%33%54%84%Specific brands

227、online store (e.g.Apple,Nike,Fenty Beauty)Membership warehouse club (e.g.Costco,Sams Club,BJs)Other online store that sells multiple brands(e.g.Best Buy,Foot Locker,Ulta)Online delivery service (e.g.Instacart,Shipt,FreshDirect,DoorDash)Other online marketplace that offers items from multiple sellers

228、 (e.g.OfferUp,Etsy,Poshmark,Wayfair)Brands cannot rely on D2C alone to scale;they will need to sell through marketplaces and other online channelsWinning brands will be those that can navigate the unique set of challenges related to multi-channel management1.“Online purchasers”are defined as adults

229、aged 18+who made an online purchase at least once in the last 12 months.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023)To scale,brands will need to pursue a multi-channel strategy,as less than 20%of online shoppers are purchasing from D2C channels ECOMMERCE

230、61ACTIVATE ECOMMERCE ENABLEMENT FRAMEWORKNote:Not exhaustive.Sources:Activate analysis,Company press releases,Company sitesPRODUCT INFORMATIONPRICINGVIRTUAL STORE CREATIONLIVESTREAM SHOPPINGVIRTUAL VISUALIZATION&FIT TOOLSSUBSCRIPTIONSCHECKOUT&PAYMENTSPOS FINANCINGFRAUD PROTECTIONCORE ECOMMERCE PLATF

231、ORMSCUSTOMER RELATIONSHIP MANAGEMENTMARKETINGSEARCH ENGINE OPTIMIZATIONFRONTEND&STOREFRONT BUILDINGCONTENT MANAGEMENT SEARCH&DISCOVERYRETAIL MEDIAEXPERIENCE OPTIMIZATION&PERSONALIZATIONDEMAND&SUPPLY PLANNINGFULFILLMENT LOGISTICSORDER TRACKING EXPERIENCECUSTOMER REVIEWS&CONTENTRETURNSRE-COMMERCECUSTO

232、MER SUPPORTLOYALTY&REWARDSDATA ANALYTICSeCommerce enablement solutions will continue to drive growth across online channelsECOMMERCEWith fast,reliable,and low-cost delivery among the top reasons why consumers purchase from online marketplaces,optimizing fulfillment will be especially critical for ma

233、rketplaces and competing online channels 621.Consumers were asked to select up to three top reasons.2.“Online marketplace purchasers”are defined as adults aged 18+who purchased from an online marketplace at least once in the last 12 months.Sources:Activate analysis,Activate 2023 Consumer Technology&

234、Media Research Study(n=4,023),Company press releases,Company sitesEXAMPLES OF FULFILLMENT LOGISTICS INNOVATION BY RETAIL COMPANIES IN RESPONSE TO HIGH CONSUMER EXPECTATIONS AROUND DELIVERYSupply Chain RoboticsAutonomous robot systems used in warehouses and fulfillment centers for tasks such as sorti

235、ng,picking,packing,and transporting to increase productivity and safetyMicro-Fulfillment CentersSmall-scale fulfillment centers positioned in densely populated areas(often within or attached to the retail companys stores),shortening the distance to customers to enable faster and more reliable delive

236、ryAlternative Last-Mile Delivery MethodsZero-occupant transportation modes such as drones,ground robots,and self-driving vehicles used to streamline last-mile delivery logistics for improved speed,cost-effectiveness,and sustainabilityFulfillment-as-a-ServiceWhite-label fulfillment solutions,enabling

237、 the retail company to monetize its supply chain network and capture valuable data to further optimize and scale its delivery operationsOne of the top reasons1 that I purchase from online marketplaces is that they offerA large selection of itemsFast,reliable,and free/discounted shippingThe ability t

238、o easily compare prices across sellersEasy and free return optionsUnique items that I cannot find elsewhereA shopping program membershipA large number of ratings/reviewsA range of easy delivery/pickup optionsThe ability to purchase new and secondhand items in one placeA guarantee to protect me again

239、st fraudulent sellersHelpful personalized recommendationsOther4%7%11%12%16%17%20%20%22%26%35%44%TOP REASONS1 FOR PURCHASING FROM ONLINE MARKETPLACES,U.S.,2023,%ONLINE MARKETPLACE PURCHASERS2ECOMMERCEEXAMPLESsupply chain robotsdelivery droneLooking ahead,online re-commerce will continue to play a key

240、 role in growing the overall eCommerce pie,with strong adoption across all age groups 631.“Online shoppers”are defined as adults aged 18+who shopped online at least once in the last 12 months.Shopping includes browsing and making purchases.2.“Likely”is defined as extremely,very,somewhat,or slightly

241、likely.3.“Online re-commerce platforms”are defined as websites/apps through which secondhand items can be purchased or sold.Sources:Activate analysis,Activate 2023 Consumer Technology&Media Research Study(n=4,023),Company press releases,Company sites,U.S.Census BureauPurchased or sold a secondhand i

242、tem online in the LAST 12 monthsLikely2 to purchase or sell a secondhand item online in the NEXT 12 months56%80%86%31%53%60%Aged 18-34Aged 35-54Aged 55+60M67M52M42M44M29MEXAMPLE ONLINE RE-COMMERCE PLATFORMS3(in beta)PARTICIPATION VS.EXPECTED PARTICIPATION IN ONLINE RE-COMMERCE BY AGE GROUP,U.S.,2023

243、,%ONLINE SHOPPERS1 BY AGE GROUPECOMMERCE 64CONTENTSPAGEConsumer Attention5Super Users10Generative AI26eCommerce52B2B and Enterprise Software64Gaming84Metaverse102Spatial Computing 118Social Video129Video135Music and Podcasts160Sports Betting177Live Entertainment and Experiences184Technology and Medi

244、a Revenues195 65Source:Activate analysis In the next 4 years,we expect global enterprise technology spend to increase by over$0.7T;this will be part of a new paradigm where buyers are increasingly focused on value,utility,ROI,and AI The way enterprises buy software is changing,characterized by:-Full

245、-enterprise decision making-Increased complexity for procurement teams-Shorter contract lengths-Time to value being prioritized over cutting-edge features In addition,the industry is shifting to meet the demands of new buyers:-Major platforms are launching AI-based features that will increase utilit

246、y-New innovators are developing point solutions to solve unaddressed pain points across both IT processes and functional business challenges This change is enabled in part by AI-based innovations(e.g.LLMs,GPU compute,connectivity)from major technology companies and a new generation of innovators The

247、se shifts will require technology companies and their customers to tackle key business challenges:-Technology companies will need to adapt their overall go-to-market approach (e.g.emphasize value and AI,update sales tactics,launch new pricing models)-Enterprise software buyers will need to ensure th

248、at they have the organization,operating model,and data strategy in place to deploy AI-based solutions at scale and ensure their software investments realize their promised ROIB2B AND ENTERPRISE SOFTWAREGrowth in enterprise IT and software will be driven by increasing demand to solve companies major

249、business problems;technology buyers expectations are significantly changing20192023E2027E$3.5T$2.8T$2.2T$1.0T$0.7T$0.4T$0.5T$0.4T$0.4T$0.6T$0.5T$0.5T$1.4T$1.2T$0.9TB2B technology spend will continue to grow and outpace total economic growth,increasing by over$0.7T globally by 2027;cloud-native appli

250、cations and services will drive the highest growth 66B2B TECHNOLOGY SPEND BY CATEGORY,GLOBAL,2019 VS.2023E VS.2027E,TRILLIONS USD1.“Cloud-native applications&services”is defined as spend on software,data storage,and compute hosted on a public cloud platform or remote data center.2.“On-premise applic

251、ations&services”is defined as spend on software,data storage,and compute hosted on-site,including servers and enterprise network equipment.3.“Devices”is defined as enterprise spend on PCs,mobile phones,tablets,and printers.4.“Third-party IT services”is defined as spend on any service offering that a

252、ssists enterprises in implementing,managing,and operating systems,software,and equipment used in modern IT environments.Does not include spend on the software,storage,and devices themselves.5.Currency neutral;represented in Aug.2023 USD.Sources:Activate analysis,Analysys Mason,Company filings,Fortun

253、e Business,Gartner,HG Insights,IBIS World,International Data Corporation,International Monetary Fund,Precedence Research,Synergy Research Group,The World BankThird-Party IT Services4Devices3On-Premise Applications&Services2Cloud-Native Applications&Services12023E-2027E CAGR:11.2%2.7%1.8%5.6%5.9%ACTI

254、VATE PERSPECTIVE In spite of market headwinds and the macroeconomic slowdown impacting technology,global spend will continue growing as companies look to deploy AI solutions at scale Spend on cloud-native applications and services will still grow at a double-digit pace;however,growth will slow down

255、compared to the 15.3%CAGR from 2019 to 2023E Technology spend growth will outpace global GDP growth over the next four years,driven by advancements in AI-based tools2019-2023E CAGR:6.4%15.3%2.4%1.0%6.6%Global GDP5 4.2%$88T$103T$117T3.1%ACTIVATE FORECASTB2B AND ENTERPRISE SOFTWAREGrowth of IT and sof

256、tware budgets are driven by companies adopting new applications and deploying them across organizations 67AVERAGE SAAS APPLICATION PORTFOLIO SIZE1 BY COMPANY SIZE,U.S.,2021 VS.2023,TOTAL SAAS APPLICATIONS1.Includes actively managed SaaS applications(i.e.actively tracked by companies,and not shadow I

257、T).2.Data represents actuals for the end of Q1 for each given year.3.Fewer than 500 employees.4.500 to 2,000 employees.5.Greater than 2,000 employees.Sources:Activate analysis,Gartner,Productiv33523847332202322021220232SMB3MID-MARKET4ENTERPRISE52021-2023 CAGR:2%2021-2023 CAGR:1

258、9%2021-2023 CAGR:22%ACTIVATE DATA PARTNERB2B AND ENTERPRISE SOFTWARE 681.“Employees”are defined as adults aged 18+who are employed full-time,employed part-time,or self-employed.Does not include military.2.“Hybrid”is defined as working a non-zero number of days both in-person and remote.3.Figures do

259、not sum to 100%due to rounding.4.Respondents indicating that they will increase or significantly increase their organizations investment in each area.Sources:Activate analysis,Activate 2022 Consumer Technology&Media Research Study(n=4,001),Activate 2023 Consumer Technology&Media Research Study(n=4,0

260、23),Gartner,National Bureau of Economic Research,OktaEMPLOYEES1 WORKING FULLY REMOTE VS.HYBRID2 VS.FULLY IN-PERSON3,U.S.,2021-2023,%EMPLOYEES1THE TECHNOLOGY CHALLENGES OF THE HYBRID WORKFORCE ARE HERE TO STAY AND WILL CONTINUE TO DRIVE SPENDTOP HYBRID WORK CHALLENGES,GLOBAL,2023,%ORGANIZATIONS23%27%

261、27%27%28%29%32%IMPROVING CYBERSECURITYSTRENGTHENING COLLABORATIONHAVING THE RIGHT TECHNOLOGYPOSITIVE WORKPLACE CULTUREEMPLOYEE PRODUCTIVITYEMPLOYEE DIGITAL PROFICIENCYEFFECTIVE ONBOARDINGPLANNED DIGITAL TOOL INVESTMENT IN THE NEXT THREE YEARS4,GLOBAL,2023,%ORGANIZATIONS63%67%69%73%73%75%80%SECURITY&

262、PRIVACYVIDEO CONFERENCINGPRODUCTIVITY&COLLABORATIONEMPLOYEE WELLBEINGEMPLOYEE ENGAGEMENTINTERNAL COMMUNICATIONPROJECT MANAGEMENTUnsolved security&collaboration challenges will increase spend202%19%20%23%23%25%62%58%55%FULLY IN-PERSONHYBRID2FULLY REMOTENote:Includes trends for total U.S.em

263、ployees1,not specific industries or functionsIT and software spend will be sustained by the need to address outstanding pain points;for example,the challenges of maintaining a hybrid workforce remain a priority for technology decision makersB2B AND ENTERPRISE SOFTWARE 69Sources:Activate analysis,CIO

264、 Insight,Gartner,Productiv,SpendfloThese challenges are so critical for businesses that purchasing enterprise technology and software is no longer just an IT problem its an organization-wide decisionACTIVATE PERSPECTIVE As part of the new software buying process,full-organization decision makers wil

265、l:Apply greater scrutiny to future software investments,ensuring that they are able to address enterprise needs Carefully weigh the promises of innovation made by new software companies against tangible ROI goals Actively monitor utilization of licensed software to identify inefficient spend Continu

266、e spending on software at high levels to maximize productivity while also prioritizing value,utility,accessibility,and sustainability goalsTHE MODERN ENTERPRISE SOFTWARE BUYING TEAM:It takes a village to select a solutionLEGAL Ensure compliance with data protection,privacy,and security regulationsFI

267、NANCE Scrutinize budget and incremental value delivered compared to existing solutionsIT Assess technology,features/functionality,and securityPROCUREMENT Evaluate vendor capabilities,negotiate MSA terms,and manage implementationLINE OF BUSINESS LEADERS Manage spend and software utilization as part o

268、f specific department P&LsEND USERS Ensure software fulfills major use cases,integrates into existing workflows,and offers necessary enablement and trainingB2B AND ENTERPRISE SOFTWARESMB1MID-MARKET2ENTERPRISE3#1Ease of ImplementationROI Within Six MonthsEase of Implementation#2Ease of UseEase of Imp

269、lementationROI Within Six Months#3ROI Within Six MonthsAbility to ScaleAbility to Scale#4ROI Within One YearEase of UseROI Within One Year#5Ability to ScaleROI Within One YearEase of Use#6Weekly Usage by All UsersWeekly Usage by All UsersWeekly Usage by All Users#7Customer Support QualityCustomer Su

270、pport QualityCustomer Support Quality#8Number of FeaturesEase of IntegrationEase of Integration#9Ease of IntegrationNumber of FeaturesNumber of Features#10Cost of SoftwareCost of SoftwareSecurity 70TOP 10 CONSIDERATIONS FOR PURCHASING SOFTWARE BY COMPANY SIZE,GLOBAL,20231.Fewer than 500 employees.2.

271、500 to 2,000 employees.3.Greater than 2,000 employees.Sources:Activate analysis,CIO Insight,Forbes,G2,GartnerEase/AccessibilityROI/ValueScalabilityFeatures of SolutionPriceA broader set of technology and software decision makers means a reshuffling of decision-making priorities;time to value,ROI,and

272、 ease of implementation are top concerns going forward80%of CIOs and CEOs experience frustration with ITs failure to deliver value73%of software buyers expressed regret about a technology purchaseMany software buyers,expecting value from their investments,experience broken promisesB2B AND ENTERPRISE

273、 SOFTWAREThe growing complexity of SaaS portfolios has increased the automation of renewal processes,and the emphasis on value has decreased the length of contracts 711.Data represents actuals for the end of Q1 for each given year.Sources:Activate analysis,ProductivSHARE OF SAAS CONTRACTS THAT ARE M

274、ANUALLY RENEWED VS.AUTO-RENEWED,U.S.,2021 VS.2023,%SAAS CONTRACTSLENGTH OF SAAS CONTRACTS,U.S.,2021 VS.2023,%SAAS CONTRACTSFewer than 6 months6 months to 1 year1 year to 2 years2 years to 3 yearsGreater than 3 yearsSoftware contracts are becoming shorter as buyers are more risk averse and can pressu

275、re sellers to agree to favorable terms333%15%67%85%An increasingly high number of SaaS contracts and growing IT complexity require automation for procurement teams to effectively manage renewals,decreasing the share of contracts manually renewedSaaS contracts that are manually renewedSaaS

276、 contracts that are auto-renewedACTIVATE DATA PARTNERB2B AND ENTERPRISE SOFTWARE202120231%1%4%6%13%15%57%58%25%20%1%4%1%6%We already see the impact of this new way of buying,as Shadow IT has fallen;nevertheless,Shadow IT still remains a significant problem for enterprises to solve going forward 72RO

277、BOTO BOLD 16PT,ROBOTO REGULAR 16 PTSHADOW IT APPLICATIONS,U.S.,2020-2023,%TOTAL SAAS APPLICATIONSACTIVATE PERSPECTIVE IT leaders are starting to address shadow IT concerns,and it will need to be a continued priority due to:Need For Visibility:Organizations require a clear line-of-sight on software u

278、tilization and performance to evaluate ROI and identify new areas for investment Security&Governance Risks:Shadow IT presents security and access management vulnerabilities that are difficult to monitor through enterprise security;54%of software buyers report purchasing software without IT team appr

279、oval2020202%59%56%52%1111Shadow IT Applications:SaaS applications that are not managed as part of a centralized enterprise contract and are instead purchased and licensed through individual employeesACTIVATE DATA PARTNER1.Data represents actuals for the end of Q1 for each given year.Sourc

280、es:Activate analysis,Gartner,ProductivB2B AND ENTERPRISE SOFTWARE20192023E2027E$1,019B$667B$377B$221B$141B$92B$327B$215B$111B$470B$311B$174BCross-Enterprise Tools4Data and Infrastructure3Individual Enterprise Applications52023E-2027E CAGR:10.8%12.0%11.1%11.2%There is high growth in demand for the fu

281、ll set of cloud-native software and services,with all applications,tools,and infrastructure forecasted to grow at a double-digit pace over the next four years 73CLOUD-NATIVE APPLICATION&SERVICE SPEND1 BY CATEGORY2,GLOBAL,2019 VS.2023E VS.2027E,BILLIONS USD1.“Cloud-native application&service spend”is

282、 defined as spend on software,data storage,and compute hosted on a public cloud platform or remote data center.2.Figures do not sum due to rounding.3.“Data and infrastructure”is defined as any spend on services that enable storing data,files,and compute capabilities on a remote system,as well as vir

283、tualization services.4.“Cross-enterprise tools”is defined as any spend on software that enhances operations across business functions,including productivity,collaboration,business intelligence,data analysis,and cybersecurity tools.5.“Individual enterprise applications”is defined as any spend on soft

284、ware that enhances specific business functions,including HR,customer service,finance,and marketing,as well as any industry-specific line-of-business tools.Sources:Activate analysis,Analysys Mason,Company filings,Fortune Business,Gartner,HG Insights,IBIS World,International Data Corporation,Precedenc

285、e Research,Statista Market Insights,Synergy Research GroupExample Leading CompaniesSTORAGE AND COMPUTEVIRTUALIZATIONAWS,Azure,Google,Snowflake,OracleDocker,Zero,Actifio,Scale Computing,VirtualBoxPRODUCTIVITYCOLLABORATIONAirtable,Notion,Atlassian,AsanaDropBox,Box,Slack,Zoom,TwilioBUSINESS INTELLIGENC

286、E AND PLANNINGCYBERSECURITYSAP,Anaplan,App Dynamics,Microsoft Power BI,DomoOkta,Cloudflare,Zscaler,Crowdstrike,JumpCloudADMIN AND HRCUSTOMER EXPERIENCEWorkday,UKG,ADP,BambooHR,ZenefitsSalesforce,Zendesk,Freshworks,HubSpot,ZohoFINANCENetSuite,Xero,Stripe,QuickBooks,RampACTIVATE FORECASTB2B AND ENTERP

287、RISE SOFTWAREINDIVIDUAL ENTERPRISE APPLICATIONSCROSS-ENTERPRISE TOOLSDATA&INFRASTRUCTUREAlongside the major technology companies,a new set of solutions is emerging to address evolving buyer needs,supported by new fundraising for AI-driven technology innovation 74Note:Not exhaustive.Includes B2B and

288、enterprise software companies that have either launched in the past two years and raised at least$10M or have had a deal size of at least$50M in the past two years.1.Robotic Process Automation.2.Includes innovators driving improvements to semiconductors,data center technology,and IoT hardware.Source

289、s:Activate analysis,Axios,Company sites,Crunchbase,Pitchbook,TechCrunchAI-ENABLED CUSTOMER EXPERIENCE/SALESCOMMUNICATIONBUSINESS INTEL.&PLANNINGAI-POWERED DEVELOPER TOOLSRPA1AI-FOCUSED CLOUD SERVICESAEROSPACE/DEFENSEBANKINGAI VIDEO/IMAGE GENERATIONECOMMERCEFINANCEHR&ADMINISTRATIONLEGAL/COMPLIANCESUP

290、PLY CHAINHEALTHCAREPUBLIC SECTORIDENTITY&ACCESSAI-BASED ANALYTICS AND MODELINGDATA STORAGEGENERATIVE AI MODELS&TOOLSNEXT-GEN HARDWARE2QUANTUM COMPUTINGNETWORK SECURITY/FIREWALLMARTECH/ADTECH/AD-OPSBIOTECH/PHARMAVERTICAL SOFTWAREAI TEXT/COPY GENERATIONAI AUDIO GENERATIONDeepLENDPOINT SECURITYTHREAT R

291、ESPONSEHORIZONTAL SOFTWAREB2B AND ENTERPRISE SOFTWARETECHNOLOGICAL BREAKTHROUGHSLARGE LANGUAGE MODELS(LLMs)DIFFUSION MODELSAI MODEL COMPRESSIONGPU CLOUD SERVICESINTELLIGENT AGENTS(RPA)5G&CONNECTIVITYCHIP INNOVATIONSQUANTUM COMPUTINGAI models trained on text-based data to understand,synthesize,and ge

292、nerate human-like text output based on predictive neural networks (vs.rules-based AI)Increasingly accessible to enterprises and end-users through API-based platforms (e.g.OpenAI,Bard)Wide applications across enterprise functions(e.g.productivity tools,communication,chatbots for IT and customer servi

293、ce)AI models able to take text-based prompts as inputs and generate images/video Enables fast creative output across a variety of use cases(e.g.advertisements,internal trainings,product customization)Image processing models have improved rapidly,with video processing models still in early developmen

294、t Reduction in size of AI models by lowering the number of parameters,compute requirements,and storage needed Requires less compute to execute,allowing for smaller chips and ability to run on edge devices Allows for more efficient model training and testing,crucial given todays scarcity of compute r

295、esources for AICloud services and analytics capabilities powered by GPU chips,able to speed up compute time and handle specific workloads through parallel compute Enables faster training of AI models,more effective forecasting and analytics,and enhanced graphics and visualization Ability to automate

296、 repetitive tasks across the enterprise (e.g.customer service interactions,data extraction,security threat detection)Developed to execute tasks based on inputs or seek and optimize for specific goalsProvides ability to transfer data across networks and between devices with higher speeds and lower la

297、tency Enables faster decision making and higher productivity for enterprises with more accurate and real-time data inputsGPU innovations enable increased edge computing power and deployment of AI capabilities,enabling real-time,localized data processing Introduction of transformer models as part of

298、next-generation chips needed for rapid training of LLMs Facilitate process improvement across operational workflows (e.g.manufacturing,forecasting,ESG and sustainability metrics)Ability to rapidly solve problems that cannot be simulated by classical computing processes (e.g.biological process modeli

299、ng,financial options pricing),enabling faster innovation and more precise decision-making capabilities Majority of the early use cases for quantum computing are for addressing public sector needs (e.g.national cybersecurity)75Sources:Activate analysis,a16z,Company sites,Forbes,The Information,TechTa

300、rget,WiredSignificant technology breakthroughs,especially in AI (e.g.generative AI,compute optimization,connectivity),will continue to be major growth enablers for the enterprise technology spaceKEY INNOVATORSB2B AND ENTERPRISE SOFTWARE 76Note:Not exhaustive.Source:Activate analysisThese technology

301、breakthroughs will drive significant productivity improvements across organizations key business functionsEXAMPLES OF INDUSTRY SHIFTS DRIVEN BY INNOVATION AND MARKET DYNAMICSEMERGING BUSINESSESKey Example:Enhancements in GPU technology enabling cloud providers to run AI-based applications at scale 3

302、NEW PRODUCT INNOVATIONKey Example:Launch of automation products simplifying workflows and boosting productivity 2NEW FEATURE DEVELOPMENT ON LARGE-SCALE PLATFORMSKey Example:Conversational chatbots enabling natural communication between users and software 1NEW RISKS AND RESPONSESKey Example:AI-driven

303、 cybersecurity products helping organizations prepare for sophisticated security threats 4SHIFTS IN IT OPERATING MODELS AND SUSTAINABILITYKey Example:Desktop virtualization and extended IT assets expanding device refurb and renewal cycles 5B2B AND ENTERPRISE SOFTWARENEW TOOLS&FEATURESVALUE DELIVERED

304、Copilot launched for Microsoft 365 as an in-program AI assistant enabled by OpenAIs GPT-4,improving Word,PowerPoint,and Excel through draft and idea creation,presentations and visuals with simple prompts,and data analysis and visualization with basic commandsImproved productivity and efficiency thro

305、ugh reduced output time for documents and analysisFreddy Copilot launched as a conversational AI-enabled tool able to automate mundane tasks(e.g.IT incident resolution,customer support analytics prompts),with specific applications for customer experience,sales,marketing,IT service management,and dev

306、eloper workflowsEnhanced employee and customer experiences through boosted productivity,automation,and specific recommendations Copilot tool for copywriters and content creators,able to take topics,articles,and style&tone recommendations as inputs to generate marketing copy,campaigns,and social medi

307、a posts able to be automatically launched through integrations Accelerated pace of marketing and media content creation,saving significant time in idea generation;dedicated prompt engineers will be important to maximize value delivery 77EXAMPLES OF INNOVATORS LAUNCHING COPILOTS AND CONVERSATIONAL TO

308、OLS1.Market capitalizations as of market close on Oct.6,2023.2.Latest valuation from Series A funding on Oct.18,2022.Sources:Activate analysis,Company sites,Crunchbase,CX Today,TechRadar,VentureBeat,Yahoo FinanceFOUNDED:1975 MARKET CAP1:$2.4TFOUNDED:2010 MARKET CAP1:$5.7BFOUNDED:2021 LATEST VALUATIO

309、N2:$1.5BConversational software experiences backed by generative AI(from both established and new companies)will drive increased ease of use,utility,and speed of onboardingNEW FEATURE DEVELOPMENT ON LARGE-SCALE PLATFORMSB2B AND ENTERPRISE SOFTWARE 781.Market capitalizations as of market close on Oct

310、.6,2023.2.Latest valuation from Series B funding on Nov.21,2019.3.Robotic Process Automation.Sources:Activate analysis,Business Wire,Company sites,Forbes,TechTargetEXAMPLES OF INNOVATORS LAUNCHING AI-BASED AUTOMATION CAPABILITIESFOUNDED:1999FOUNDED:2003FOUNDED:2003MARKET CAP1:$202BMARKET CAP1:$114BL

311、ATEST VALUATION2:$6.8B NEW TOOLS&FEATURES LAUNCHED IN 2023:Salesforce Flow launched as an automation tool for the Salesforce platform,allowing users to orchestrate workflow processes with specific flowcharts and automate repetitive tasks (e.g.email outreach,manual customer data entry)with Flow RPA3

312、Newly launched Utah platform showcasing ability to improve workflows with Process Automation tools and team efficiency with Workforce Automation toolsLaunched Automation Copilot and Generative AI for Automators functions,allowing users to create orchestration of complex tasks (e.g.reviewing of expen

313、se and invoices,document revisions)through translation of simple language and stepsEXAMPLE USE CASES:Customer Engagement:Orchestration of specific responses to CRM data triggers to deliver personalized outreach Communication and Collaboration:Integration with Slack triggers responses based on messag

314、esWorkforce Efficiencies:Visibility into employee productivity and forecasting for staffing demand requirements with dynamic scheduling IT Incident Resolution:Mapping of IT operational processes to identify causes of incidents and speed up future processesAdmin and HR Tasks:Automation of time-intens

315、ive back-office processes including expense management,accounts payable,and tax compliance Finance Operations:Efficiency and error prevention for accounting and invoicingTechnology platforms are launching AI-based automation features that improve workflow productivity and orchestration across both h

316、igh-volume simple tasks and complex processesNEW PRODUCT INNOVATIONB2B AND ENTERPRISE SOFTWARECreated161empty127Draft140Pending Dispatch140Hi!Tell me what you would like to automate today.14:38|3/30/2314:38|3/30/23MarcusCo-PilotI need an automation that can read a list of automotive parts from a pdf

317、,and pulls the parts description and pricing from auto supplier websites.Highlight the best price and email the collected information.StateAssignment GroupProcess StartProcess EndsConnectionsActivitiesGPU CLOUD COMPUTE PRICING1,U.S.,2023,USDLEGACY CLOUD PROVIDERSAI-FOCUSED CLOUD PROVIDERS 79Note:Exa

318、mples are not exhaustive.1.Pricing based on the hourly cost of compute for an on-demand NVIDIA A100 40GB.Sources:Activate analysis,a16z,Bloomberg,Company press releases,Company sites,The Information,Precedence Research,S&P Global,Semiconductor Industry Association,TechCrunch$2.93$3.05$3.40$4.10$1.10

319、$1.73$2.06END APPLICATIONProduce contracted semiconductors forSupply chips to power the data centers ofProvide the compute for the training and workloads ofAre refined and optimized byDeliver a use-case tailored model forProvide API-driven access forCLOSED-SOURCEOPEN-SOURCEOPEN-SOURCE DERIVATIVESSEM

320、ICONDUCTOR FABRICATORSCHIP DESIGNERSCLOUD PROVIDERSGENERATIVE AI MODELSMODEL TRAINING TOOLSEXISTING AI SOFTWARE LAYERGrowth driven by increasing complexity of chip design and manufacturing and accelerated public sector funding(e.g.CHIPS and Science Act)Growth driven by necessity of highly efficient

321、GPUs/TPUs for model training workloads,facing scarcity todayGrowth driven by emergence of specialized AI-focused cloud providers alongside legacy cloud providers,offering more efficient GPU compute highly tailored for AI workloads,delivering more attractive cost of compute compared to running the sa

322、me workloads on CPUsGrowth driven by development of increasingly sophisticated model capabilities,with ongoing adoption across enterprisesGrowth driven by requirements for companies to enhance and tailor models to address specific use cases and data sets Growth driven by software companies building

323、AI capabilities through existing tools,API access,and easy to use functionality2027E Market Size:$1.0T2027E Market Size:$170B2027E Market Size:$220B2027E Market Size:$60BNew AI capabilities are giving rise to a new cloud compute and storage ecosystem focused on GPU compute and solutions to access AI

324、 infrastructure(e.g.generative AI models,training tools)EMERGING BUSINESSESB2B AND ENTERPRISE SOFTWAREMAJOR CYBERSECURITY CHALLENGESRESPONSES FROM INNOVATORSAI-POWERED CYBERATTACK VULNERABILITY Potential risk for AI to be used as a force-multiplier in cyberattacks in the future Sophistication of thr

325、eats will be increased by generating malicious code,reverse-engineering security offerings,and complex phishing FOUNDED:2011 MARKET CAP1:$42BCounter Adversary Operations product launched to actively monitor and track potential AI-based threats,leveraging AI trained on network events and signals to s

326、pot irregularitiesCLOUD SECURITY THREATS Data breaches caused by poor cloud management,unauthorized access,and insecure data are often unnoticed due to a lack of visibility Breaches create data and privacy compliance challenges (e.g.localization requirements)FOUNDED:2020 LATEST VALUATION2:$10BWiz In

327、tegration Platform released to plug into and observe all cloud access points,monitoring users,data,and workloads to mitigate risk as customers scaleENDPOINT SECURITY RISKS Network-connected devices are more vulnerable to security threats than core networks Remote working environments and shadow IT a

328、pplications heighten endpoint security risks FOUNDED:2007 LATEST VALUATION3:$9BSBOM for Comply launched to give full visibility into software and endpoint supply chains(including devices and software licenses),ensuring proper use and access 801.Market capitalization as of market close on Oct.6,2023.

329、2.Latest valuation from Series D funding on Feb.27,2023.3.Latest valuation from funding round on June 26,2020.Sources:Activate analysis,Business Wire,Check Point Software,Company sites,Crunchbase,CSO Online,Forbes,IBM,PitchbookAcceleration of enterprise technology and software complexity presents cy

330、bersecurity risks;in response,innovators are launching new products to address these threatsEXAMPLES OF INNOVATORS LAUNCHING CYBERSECURITY TOOLSNEW RISKS AND RESPONSESB2B AND ENTERPRISE SOFTWAREACTIVATE PERSPECTIVE The continued adoption of and advancements in virtualization will drive:Lower require

331、ments for hardware performance and refresh,as virtualization will allow users to access the same compute capabilities on used devicesA focus on IT refurb and renewal,increasing the sustainability of IT operating processesGreater investments in the circular IT economy by hardware companiesExample Cir

332、cular IT ProgramsEXAMPLES OF MAJOR VIRTUALIZATION SERVICES(e.g.Cloud PCs,Next-Gen Virtual Desktops)Provides cloud-based access to Windows interface and software on any device,improving efficiency of IT through faster onboarding and setup,ease of access,and less reliance on hardwareEnables virtual de

333、sktops for users with specific software installations and user permission,allowing for quick onboarding and customization for employees 81Sources:Activate analysis,Company sites,Gartner,GeekflareENABLERS AND IMPACT OF VIRTUALIZATION SERVICESENABLERS OF VIRTUALIZATIONIMPROVED CONNECTIVITYHigher speed of data transfer enables more devices to be run off of one networkIMPROVED COMPUTEStrong compute c