《埃森哲：2024高科技领域的生成式AI机遇报告：学习引领并行不悖（英文版）（18页）.pdf》由会员分享，可在线阅读，更多相关《埃森哲：2024高科技领域的生成式AI机遇报告：学习引领并行不悖（英文版）（18页）.pdf（18页珍藏版）》请在三个皮匠报告上搜索。

1、High techs Gen AI opportunityLearn and lead at the same time.Even as high tech companies adopt Gen AI technologies to increase efficiency and fuel creativity within their own organizations,they must alsosimultaneouslyenable customers Gen AI experiences and drive new revenue creation.Accomplishing th

2、ese goals wont be easy.Accentures point of view offers a high-level guide for companies on this journey.2High techs Gen AI opportunityIn March 2023,Accenture published a report called“A new era of generative AI for everyone.”1 In it,we noted that ChatGPT,which reached 100 million monthly active user

3、s just two months after launch,was the fastest-growing consumer application in history.2 Moreover,our research found that 40%of all working hours could be impacted by Large Language Models(LLMs).3 Since then,the pace at which businesses have started to explore Gen AI has surpassed even the most opti

4、mistic expectations.For example,more recent Accenture research has found that 73%of high tech CxOs are interested in exploring the potential use of applications like ChatGPT for tasks such as natural language processing,text generation and customer service automation.4 Most,if not all,business leade

5、rs are either investing in Gen AI or considering its potential to enhance or transform ways in which their companies work and innovate.of our largest 300 clients are applying Gen AI in some capacity.However,they understand that scaling will be a problem that transcends the technology itself and requ

6、ires upskilling people in order to apply it.80%3High techs Gen AI opportunityHigh tech companies have a unique role in this Gen AI transformation.They must,like all enterprises,adopt Gen AI within their own organizations to innovate and be more efficient.The technology will drive change across opera

7、tions,including finances,supply chains,manufacturing,customer success and talent.Yet,high tech companies will also supply the means to enable this technology.High tech offeringsin the form of semiconductors,smartphones,PCs,high-end medical equipment,servers or other enterprise productsare the founda

8、tion for customers across industries accessing the potential of this new technology.How high tech companies embed Gen AI in their products and solutions will direct how it is used by companies in other industries.Ultimately,high tech companies must learn and lead at the same time.Each step they take

9、 as Gen AI adopters will increase their ability to enable Gen AI transformation across all industries through their products.And vice versa.How well they fulfill the roles of adopter and enabler will have far-reaching effectson their own employees,on their customers and,ultimately,on end-users.Learn

10、 and lead in the Gen AI transformationHow high tech companies embed Gen AI into their products and solutions will direct how it is used by companies in other industries.4High techs Gen AI opportunityAdopting Gen AI a catalyst for change“Be the change you wish to see in the world.”Its well-known as a

11、n inspirational quote;it should also become the mantra for high tech companies as they adopt Gen AI.5High techs Gen AI opportunityLeaders in high tech are already aware of market shifts pointing to a need to enhance their customers experiences through as-a-Service(aaS)offerings.These offerings lower

12、 the barrier to entry and deliver stable sources of recurring revenue.Their design also allows for more proactive and personalized customer engagement.Building an organization to deliver on these offerings is not easy,and many high tech companies have not yet made the organizational changes necessar

13、y to respond to these market shifts.Gen AI could be the game changer here.Intelligent adoption of Gen AI capabilities across the organization can supercharge the shift towards building for aaS offerings.It works at the nexus of hardware and software to build out capabilities at speed.It allows R&D t

14、eams,for example,to capture deep customer insights,quickly iterate prototypes and improve product quality.It delivers deeper product usage insight and advice,enabling a customer success model for working with clients by proactively handing issues.It can connect across multiple data sets to mitigate

15、supply chain risks and drive impactful and personalized marketing communications.These create a conducive environment to offer higher risk/reward models and take complications of managing platforms away from their customersall at a reasonable price,with greater reliability and higher margins.Put ano

16、ther way,Gen AI technology is a spark that can set off powerful organizational change.A high tech companys effective use of Gen AI today makes it a“poster child”for the vision of what its products can enable for other companies.What might this look like in practice?It could be as straightforward as

17、investing in Gen AI infrastructure inside of their organizations to accelerate the work they do today(say,improving their capacity to develop software).Data scientists and AI practitioners already spend over three-quarters(77%)of their time on data pre-prep,training,modeling and tuning,tasks that re

18、quire huge amounts of GPUs,memory and power.5 An upfront investment in this essential infrastructure can deliver significant cost savings and revenue growth over time.The cash infusion companies could experience may be used as support for long-term strategies to get closer to customers.It could also

19、 be as interesting as using Gen AI-enabled services to aid in launching a new product or service,by bridging interconnected and cross-functional roles,people(R&D,sales,marketing,customers,service)and responsibilities.Gen AI is a game changer that can supercharge the shift towards building for aaS of

20、ferings.6High techs Gen AI opportunityThe accompanying graphic describes and maps six use cases that offer high tech companies the chance to show how Gen AI can transform organizations from within.6 The guiding principle for Gen AI adoption in high tech companies as counter-intuitive as it may seemi

21、s putting the people first.That means designing Gen AI solutions for the organization with continuous input from the people who will be using them,ensuring the technology reflects and enables peoples ingenuity while making space for mistakes and for learning.It means designing for accessibility to h

22、elp all employees contribute to their full potential.7 And,finally,it means calibrating trust,so that everyone has the confidence to use Gen AI without fear.The combination of diverse users nuanced and creative perspectives will serve to improve the technologys interface.Bottom line:adopting Gen AI

23、is modeling what high tech companies seek to enable.How the industry adopts this technology will shape how they enable it across industries.7High techs Gen AI opportunityMarket TrendsSemiconductorProduct DesignManufacturing Semi ElectronicsProduct Development and ProductionDistribution&After SalesCh

24、annel PartnerConsumerEnterpriseNetworkConsumerMedTechValue chain transformation:six areas of opportunity Implementing personalized recommendations to boost low-touch salesProviding field technicians with methods to expedite onboarding and reduce errorsImproving supply chain resiliency through Materi

25、al Risk Management Deliver deep insights to customer support,success and downstream organizationsEnhancing OEE from initial production through Visual Inspection Rapid prototyping and product design for newmarket segments Visual Inspection using synthetic data to improvedelivered quality Cross-sell a

26、nd upsell with complex solutionrecommendations with virtual assistants Persona-based marketing material generated fromexisting documentationUtilizing scenario modeling to enhance financial resilience Generate real-time what-if scenarios withcross-organizational collaboration Identify variances and r

27、educe time to generatereports and narratives Context based information for tech to completework fast and accurate Create self-service portals with Gen AI-assistedknowledge management Infuse LLMs for due diligence and end-2-endcontract management Improve visibility and accurate recommendationson sour

28、cing pathway Improve CLV by guiding customer successmanagers on client discussions Deliver insights and guidance to SMB partnerswith impactful narratives1354628High techs Gen AI opportunity531246Enabling Gen AI:Collaborate more than everAccentures Chief Technology and Innovation Officer,Paul Daugher

29、ty,as he was referring to the use of OpenAI to generate a working website from a single scrap-paper drawing.And that sums up the potential of Gen AI enablement:empowering great ideas by facilitating their execution.“The hottest new programming platform is the napkin.”9High techs Gen AI opportunityTo

30、 become a successful enabler of Gen AI technology,high tech companies cannot do it alone:they need an ecosystem-based approach.Reliably developing Gen AI offerings is a challenge.Difficulties in finding funding,accessing raw material,building infrastructure and more make it difficultand riskyfor any

31、 one company to move on its own.By building out an intentional ecosystem,companies can leverage their strengths and hedge against risks,while still bringing tailored solutions to customers faster and at scale.The ecosystem solution facilitates and accelerates collaboration among partners,including c

32、hip makers,fabs,server/network providers and co-location centers.Consider what a developer community might need to build solutions that are hardware agnostic and offer cost-effectiveness from the outset.The UXL Foundation(Intel,Qualcomm,Google and others)has addressed that need with an open ecosyste

33、m assembled to accelerate AI workloads with an abstraction layer that enables developers to build transferable solutions across hardware architectures.8 By contrast,the risks of going alone can be enormous.Think back to the telcos that spent more than$500 billion building broadband networks from 199

34、6 to 2002.9 With the internet still in its infancy,demand grew slowly and they had no way to recoup their investments.The result was$2 trillion in losses to shareholders,thousands of workers laid off and several major bankruptcies.10 Roll on a few years and many software and platforms companies took

35、 advantage of that excess and unused capacity,reaping profits in the process.High tech companies cannot go it alone:they need an ecosystem-based approach.10High techs Gen AI opportunityTo lead an ecosystem solution requires thinking about who needs to come with youand about who will come along after

36、 you.Strong,equitable ecosystem management and governance from the start,led by high tech players,is essential,bearing in mind that the winning solution may even require ecosystem players to disaggregate their products and services to work effectively with partners.The accompanying chart shows the k

37、ey elements of an ecosystem along the development of a Gen AI capability,showing why going alone isnt a financially or commercially viable option.As the figure shows,there are still relatively high barriers to entry if a company wants to go at it alone.Some entry points are still too costly or scarc

38、e.Others demand huge amounts of capitalassuming the pieces are producedin terms of dollars and time.An ecosystem approach offsets both barriers and costs.Building a strong ecosystem around Gen AI offerings is about developing a long-term vision for this technology.11High techs Gen AI opportunityFund

39、ingSilicon Servers Foundation Models ApplicationsThere are high barriers to entry if a company is going at it alone.Some entry points are still costly or scarce.An ecosystem approach offsets both barriers and costs.VC Gen AI deals declined 29%q/q but ended higher at$6.1B due to Amazons almost$4B dea

40、l with Anthropic.11Big tech companies committed to invest in AI.15Gen AI continues to see investment as VC and PE dry powder is worth trillions of dollars in the aggregate(theres money on the table).Chip shortages are producing winners and losers in the AI gold rush(“Magnificent 7”vs academia).To of

41、fset the chip shortage,semiconductor companies can focus on innovative packaging while promoting accessibility.Microsoft has committed to invest$13B in OpenAI.12Server shipments are falling while cloud demand is rising as AI costs are delaying server refresh cycles.13Forecasts for data-center server

42、 growth far exceed any that occurred during the rise of the internet but cannibalization may occur.16 The massive power outlay creates an opportunity to explore new ways to power data centers;renewables and ESG is a win-win for all parties.AH estimates for a GPT-3 single run training cost ranges fro

43、m$500K to$4.6M depending on hardware assumptions.14 GPT-4 is the new space race.The“real”AI brick wall is inference.The goal is to decouple training compute from inference compute.17 Infrastructure challenges creates an opening to invest in research with smaller partners(academia,startups)to make ou

44、tcomes practical.The infrastructure costs for Gen AI chat bots can be billions of dollars for hyperscalers.The margins for AI applications are significantly smaller than other SaaS solutions.18To explore new applications,high tech companies should develop consortiums that can help lower the barriers

45、 to entry for enterprises.The ecosystem advantage:key elements Source:Accenture Research12High techs Gen AI opportunityHigh tech ecosystems are already forming Were already seeing high tech companies bringing together the type of ecosystems that will be vital in a Gen AI world.Consider the way Lenov

46、o has expanded the availability of its AI-ready smart devices and edge-to-cloud infrastructure to include new platforms purpose-built for enabling AI workloads.19 Or NVIDIAs AI Workbench,which allows developers to quickly create,test and customize pretrained Gen AI models in just a few clicks and th

47、en scale them to virtually any data center,public cloud or NVIDIA DGX Cloud.20 In another example,Intel collaborated with Accenture to build downloadable AI reference kits to the open source community.21 The goal is to help enterprises accelerate their AI deployments in areas like image and voice ge

48、neration and natural language semantic searchwhile potentially reducing running costs.Or HPE,which is broadening its GreenLake portfolio,partnering with the German AI company Aleph Alpha to offer GreenLake for large language models.Customers can build their own AI models using their own data,without

49、 paying for or renting the supercomputers needed because the offering runs on HPEs own Cray XD supercomputers,which have an AI-native architecture.22According to Accentures Tech Vision 2023 report,97%of global executives agree that AI foundational models will enable connections across data types,rev

50、olutionizing where and how AI is used,and 98%agree that AI foundation models will play an important role in their organizations strategies in the next three to five years.2313High techs Gen AI opportunityFive fundamentals of Generative AIAs our clients embark on their generative AI reinvention,there

51、 are five key imperatives that they need to invest in.These are key to driving success.Reinvent talent and ways of working Adapt operating models for new ways ofworking.Cultivate a corporate culture thatplaces talent development at the forefront.Commit to continuous learning and skillingto maximize

52、the potential of AIespeciallybecause of the changes that happen rapidlywith the technology.Close the gap on responsible AI Develop and integrate a strategy for responsibleAI that brings intentional actions to design,deploy,and use AI to drive value whileprotecting from the risks of AI.This includes

53、the strategy and developmentof responsible AI monitoring and compliance,as well as employing AI security into theirentire value chain.Understand and develop an AI-enabled secure digital core Data is for AI,therefore enabling a moderndata platform is very important.Re-architecting applications to be

54、AI-readyis also critical,with a flexible architecture thatallows you to access to a range of models inpartnership with the ecosystem.We build thiswith security as a foundational element.Drive continuous reinvention Embrace a modular,step-by-step approachto innovation that spans multiple years.Cultiv

55、ate a corporate culture that viewscontinuous reinvention not just as a strategy,but as a core competency.Gen AI led reinvention journeys are complexand multi-year in nature.Companies need tobe prepared to allocate the capital,time andtalent for the journey.Lead with value Shift the focus from siloed

56、 use-cases toprioritizing business capabilities acrossthe entire value chain based on return oninvestment.Clients need to be value-led inevery business capability that they chooseto reinvent with Gen AI.We recommend companies pursueinvestments in two categories:Table stakesinvestments that offer rad

57、ical efficiency(but will also soon become the new industrystandard way of operating)and strategicbets that offer truly novel advantage.14High techs Gen AI opportunityLeading the Gen AI revolutionBy embracing Gen AI as adopters and enablers with thoughtfulness and intention,high tech companies will h

58、ave the power to revolutionize not only their own operations but also the lives of their customers and the consumers they serve.They will become the driving force behind the future of AI and the global compute infrastructure that supports it.In doing so,they will position themselves as indispensable

59、 players in the most significant technological revolution since the advent of the Internet.Gartner found by 2026,more than 80%of enterprises will have used generative artificial intelligence(GenAI)application programming interfaces(APIs)or models,and/or deployed GenAI-enabled applications in product

60、ion environments,up from less than 5%in 2023.24Intels CEO,Pat Gelsinger,has called the converged world of smart connected devices and human experience in which we live and work today the“siliconomy.”25 Now,with the incredible advances were seeing in LLMs and in Gen AI more generally the siliconomy i

61、s entering a new era.Its set to power an exponential leap forward in growth and productivity.High tech companies have an unprecedented opportunityand an existential responsibilityas part of this Gen AI revolution.Leading in Gen AI means walking joyfully into an unknown future.Success will require a

62、deliberate and dedicated strategy,built on this continual process of adoption and enablement,to come out ahead.We believe that high tech industry is ready to take those steps.After all,theyre the very foundation of Gen AI.And the siliconomy itself.15High techs Gen AI opportunity1.Accenture,“A new er

63、a of generative AI for everyone”,Accenture,March 2023 https:/ Hu,“ChatGPT sets record for fastest-growing userbase”,Reuters,February 2,2023 https:/ new era of generative AI for everyone”,Accenture,March 2023 https:/ Enterprise Reinvention”,CxO global survey of2325 C-Suite executives,n=91 high tech r

64、espondents,data collected January February 20235.Zeeshan Hussain,“AI Infrastructure is Booming“,E,September 8,2023 https:/ Vision 2023 for High Tech WhenAtoms meet Bits”,Accenture,2023 https:/ Henneborn,“Designing Generative AI to Work for People with Disabilities”,HBR,August 18,2023 https:/hbr.org/

65、2023/08/designing-generative-ai-to-work-for-people-with-disabilities8.The Linux Foundation,“Unified Acceleration Foundation Forms to Drive Open Accelerated Compute and Cross-Platform Performance”,September 2023 https:/www.linuxfoundation.org/press/announcing-unified-acceleration-foundation-uxl9.Robe

66、rt E.Litan,“The Telecommunications Crash:What To DoNow?”,Brookings,December 1,2002 https:/www.brookings.edu/articles/the-telecommunications-crash-what-to-do-now/10.Dale A.Oesterle,“The year of the Telekom Meltdown“,2002 http:/www.jthtl.org/content/articles/V2I1/JTHTLv2i1_Oesterle.PDF#page=311.Nasdaq

67、,“LatAm Tech Weekly:AI VC Deals Ease up,Backlog ofTech IPOs,October 2023 https:/ billion bet on OpenAI carries hugepotential along with plenty of uncertainty”,April 2023 https:/ 13.The Register,“Another thing you can blame AI for:Cloudgrows but server shipments are down”,August 2023 https:/ Appenzel

68、ler,Matt Bornstein,and Martin Casado,“Navigating the High Cost of AI Compute”,August 2023 https:/ Press,“Amazon is investing up to$4 billion in AIstartup Anthropic in growing tech battle”,September 2023 https:/ 16.MarketWatch,Unprecedented AI server buildout may behappening too quickly,this analyst

69、warns”,October 2023 https:/ 17.Semianalysis,“GPT-4 Architecture,Infrastructure,Training Dataset,Costs,Vision,MoE”,July 2023 https:/ 18.PYMNTS,“Is Going for Broke With ChatGPT Bankrupting SamAltmans OpenAI?”,August 2023 https:/ Smith-Goodson,“The Extraordinary Ubiquity Of Generative AI And How Major

70、Companies Are Using It”,July 21,2023 https:/ AI Workbench Speeds Adoption of Custom Generative AI for Worlds Enterprises“August 8,2023 https:/ on the promise of AI everywhere”November 2023 https:/ Wheatley,“HPE debuts GreenLake for LLMs,anadvanced cloud for generative AI workloads”,June 20,2023 http

71、s:/ Vision 2023”,January 2023 https:/ 24.Gartner press release,“Gartner Says More Than 80%ofEnterprises Will Have Used Generative AI APIs or Deployed Generative AI-Enabled Applications by 2026”,October 11,2023 https:/ GARTNER is a registered trademark and service mark of Gartner,Inc.and/or its affil

72、iates in the U.S.and internationally and is used herein with permission.All rights reserved.25.Intel,“Intel Innovation 2023”,September 20,2023 https:/ techs Gen AI opportunityPadampreet SMatthew J.HHarmandeep AHarish NAuthors17High techs Gen AI opportunityAbout AccentureAccenture is a leading global

73、 professional services company that helps the worlds leading businesses,governments and other organizations build their digital core,optimize their operations,accelerate revenue growth and enhance citizen servicescreating tangible value at speed and scale.We are a talent and innovation led company w

74、ith 733,000 people serving clients in more than 120 countries.Technology is at the core of change today,and we are one of the worlds leaders in helping drive that change,with strong ecosystem relationships.We combine our strength in technology with unmatched industry experience,functional expertise

75、and global delivery capability.We are uniquely able to deliver tangible outcomes because of our broad range of services,solutions and assets across Strategy&Consulting,Technology,Operations,Industry X and Accenture Song.These capabilities,together with our culture of shared success and commitment to creating 360 value,enable us to help our clients succeed and build trusted,lasting relationships.We measure our success by the 360 value we create for our clients,each other,our shareholders,partners and communities.Visit us at Copyright 2023 Accenture.All rights reserved.