《谷歌云（Google Cloud）：2023年数据和 AI 趋势报告（英文版）（43页）.pdf》由会员分享，可在线阅读，更多相关《谷歌云（Google Cloud）：2023年数据和 AI 趋势报告（英文版）（43页）.pdf（43页珍藏版）》请在三个皮匠报告上搜索。

1、2023Data and AITrendsReport2023 Data and AI Trends ReportFive trends amounting to an interconnected data strategyShow data silos the door01Usher in the age of the open data ecosystemEmbracethe AItipping pointInfuse insights everywhereGet to know your unknown data202030405Youll find some common denom

2、inators across the trends,including the overarching need for greater unification and flexibility.Youll also learn why these trends depend on each other for success.2023 Data and AI Trends ReportIntroduction3Youre contributing to the fastest progression of innovationandare managing more changethanany

3、 generation before you.As you look at the five data and AI trends in this report,the way they are evolving and how global organizations are contributing to them could surprise you.Thats because our challenges today are different than they were a year ago.Consumer demand,market conditions,and new AI

4、and machine learning technologies have evolved.So has our perspective.Were all managing increased data complexity,looking for new patterns,creating new models,making data available to the right people and applications at the right time,and keeping track of every byte in a way that meets requirements

5、.To identify the current data and AI trends,we partnered with IDC on multiple studies involving global organizations across industries.We then asked Google thought leaders to weigh in on the research and reveal whats most important for organizations data and AI strategies.2023 Data and AI Trends Rep

6、ort4Show data silos the door01Show data silos the doorBy 2026,82%of organizations are looking to ensure that all capabilities supporting the full data and AI workflow are tightly integrated in their cloud data platform.1 2023 Data and AI Trends Report5Show data silos the doorTrend 1Show data silos t

7、he doorA unified data cloud provides a platform that supports every stage of the data lifecycle.Databases,data warehouses,data lakes,streaming,BI,AI,and ML all reside on a common infrastructure that is pre-configured to work together seamlessly.More efficient data usage and accessibilityAccelerate d

8、ecision making and development cyclesImprove customer experience2023 Data and AI Trends ReportTrend 16Show data silos the door0102032023 Data and AI Trends Report7Show data silos the doorTrend 1Andi GutmansGM and VP of Engineering for Databases,Google CloudOrganizations are realizing that their silo

9、ed data storage and warehouse strategies cant keep up with modern demands.With the amount of data that devices and applications generate every day,it doesnt surprise me.They need a better way to store,manage,analyze,and govern all this data,while also cutting down on the extra work,costs,and conflic

10、ting insights caused by silos and redundant systems.The skills of developers,IT administrators,security analysts,and business teams are best used developing innovative applications and bringing services to market faster,not chasing after data.These contributors must know what data exists and where i

11、t lives and be able to easily access and analyze up to date data.With a unified data cloud,this all becomes possible.For me,one of the biggest takeaways about this trend is that a unified data cloud enables the integration of data and insights into digital experiences and workflows.And as a result,u

12、sers can have the right information,exactly when they need it,to reach the best possible outcomes.“generated each year is original,while the remaining 90%is replicated.2By 2026,7 PB of data will be generated per second globally.At the same time,only 10%of the data2023 Data and AI Trends Report8Show

13、data silos the doorTrend 1How industries are taking advantage of a unified data cloudRetailMore retailers are bringing all their data into one platform to get the customer insights they need to deliver a unique,personal experience across all consumer touchpoints,increasing customer loyalty and conve

14、rsion rates across all channels.ManufacturingManufacturers are bridging connectivity among disparate machines and systems with a unified data platform,making their data easier to use which strengthens their connections with suppliers and lets them act quickly to prevent shipping delays.Financial Ser

15、vicesRetail banking and insurance firms are using privacy and customer-centric data solutions to enable better personalization,more effective marketing analytics,and customized direct-to-customer experiences.2023 Data and AI Trends Report9Show data silos the doorTrend 1What can your organization ach

16、ieve with a unified data cloud?Equifax broke down 80-plus data silos into a seamless data fabric,giving them the ability to respond faster to customer and regulatory needs.AmericasEuropeAsiaDelivery Hero integrates data into a unified data platform for cost savings and better customer experiences.Sh

17、areChat simplifies management and prepares the company for growth with a unified,optimized infrastructure.“As we expand,well build new algorithms that process real-time datasets in regional languages and accurately predict what content users want to see.Google Cloud gives us an infrastructure optimi

18、zed to handle such compute-intensive workloads for current and future growth.”Bhanu Singh,Co-founder and Chief Technology Office,ShareChatWhen operational and analytical systems are decoupled,organizations struggle to piece together different solutions to build intelligent,data-driven applications.T

19、o meet customer expectations and deliver“always on”digital experiences,operational and analytical systems need to work together on the same data in near real time.2023 Data and AI Trends Report10Usher in the age of the open data ecosystem02Usher in the age of the open data ecosystem78%of executive m

20、anagement believe that using external data is a critical competency for their enterprise.3Trend 22023 Data and AI Trends Report11Usher in the age of the open data ecosystemIntegrate data with your technologies of choice while avoiding lock-inIncrease ROI of existing investmentsFaster development cyc

21、les0102032023 Data and AI Trends Report12Usher in the age of the open data ecosystemUsher in the age of the open data ecosystemTo protect technology choice and reuse code and standards-based services,more organizations are adopting open source software and open APIs.Trend 2Trend 2Gerrit KazmaierVP a

22、nd GM,Data&Analytics,Google Cloud2023 Data and AI Trends Report13Usher in the age of the open data ecosystem“is the most important aspect to emerging AI adoption strategies.”Unlocking that data by building anopen,multicloud data ecosystemIncreasing requirements for data ecosystem flexibility are tak

23、ing conversations about open standards,data integration,and technology choice to new levels.Organizations recognize that their data is at the heart of digital innovation,and its the key to unlocking AI.The challenge is that data is being generated at far greater rates than ever before,and its being

24、trapped within the new silos of different point solution formats and closed clouds.Unlocking that data by building an open,multicloud data ecosystem can improve everything.This can lead to faster time to market,and improved return on investment.More importantly,it can make your organization more com

25、petitive.Imagine,all your employees,customers,and partners taking part of your data ecosystem as contributors rather than bystanders.Organizations want the freedom to create a data cloud that includes all formats of data from any source or cloud.They want to use the technologies that work best for t

26、heir specific needsand to increase the pace of innovation without having to worry about technology silos and debts.This is ultimately about unlocking the power of data and AI for all companies out there.”“Trend 2Opening systems to allow for data movement and multicloud analytics2023 Data and AI Tren

27、ds Report14Usher in the age of the open data ecosystemTraditionally,organizations deployed individual systems and tools to resolve specific problems.As a result,many organizations now store data across multiple platforms and public clouds.Often,this data ends up being siloed,making it hard to get in

28、sights across all of the data.The adoption of open standards and open architectures helps companies avoid lock-in and silos by protecting the freedom to move data between platforms as needed to support workflows,insights,and data monetization.For example,data stored in any SQL-based relational datab

29、ase such as PostgreSQL can be easily moved and shared with any other SQL-based database.Systems with open APIs that conform with the REST architectural design make it easy for companies to consume and share data from internal and external sources.At the same time,the use of open standards and open a

30、rchitectures also enables organizations to minimize data movement and egress fees by analyzing data where it lives.70%more data leaders than data laggards say its important for data clouds to be based on open protocols.4Trend 2Open to open sourceResearch shows that open source adoption is increasing

31、,while the use of licensed enterprise software is decreasing.Highlights70+%of new apps will be developed on open source databases 80%of enterprises will be multicloud open source is critical for flexibility“Cloudification”of open source databases with fully managed services is considerably growing i

32、n market size.Open Source LicensePopularityCommercial License“Reuse is a foundational engineering principle for improving productivity.Open ecosystems go back to that principle by using open standards and open source technologies to make data,code,and applications discoverable and portable but also

33、protected by one consistent security layer.”Firat Tekiner,Senior Product Manager,Google Cloud2023 Data and AI Trends Report15Usher in the age of the open data ecosystemOpen source software is playing a key role in data ecosystems:Organizations are speeding up development and lowering costs by using

34、pre-built,pre-tested open source services and applications such as PostgreSQL,Kafka,TensorFlow,PyTorch,Presto,JanusGraph,and Apache projects.For example,organizations are building their data lakehouse using open source technologies,storing data in open formats like Apache Parquet with processing eng

35、ines like Apache Spark,and using open frameworks like Apache Iceberg and Delta.Cloud providers open-source-as-a-service offerings give companies the freedom to adopt open source software while enjoying the backing and expertise of dedicated engineering resources.5Trend 2Open for more dataModerna fin

36、ds new ways to help people with mRNA by leveraging and integrating the best technologies available.AmericasEuropeAsiaSwisscom increases visits by 25 percent by integrating geo targeting,responsiveness,and website dataTokopedia runs their ecommerce platform on Kubernetes to improve experience and kee

37、p shoppers coming back.“Looker fits well with our multicloud philosophy because we can choose our preferred database and leverage integrations to make our data accessible and actionable.Overall,Google is making a lot of progress in multicloud,which allows you to not have to think about the vendor an

38、d just adopt what you need to do the job well.”Dave Johnson,VP of Informatics,Data Science,and AI at Moderna2023 Data and AI Trends Report16Usher in the age of the open data ecosystem78%of executive management believe that using external data is a critical competency for their enterprise.6Organizati

39、ons are making use of the publicly available datasets such as weather,trend,and location data to extract valuable insights and develop revenue generating applications.And today,75 percent of organizations are using location data across a broad range of business functions and processes,7 such as supp

40、ly chain,public transportation,and personalized customer experiences.Not only are public datasets available on demand but theyre also free of administration and maintenance costs,and vetted by the community for accuracy.Teams can also further expedite data pipeline development when they can access p

41、ublic data sets via open standards-based APIs and follow consistent standards for consumption and ingestion.2023 Data and AI Trends Report17Embrace the AI tipping point03Embrace the AI tipping pointBy 2025,at least 90%of new enterprise application releases will include embedded AI functionality.8Tre

42、nd 32023 Data and AI Trends Report18Embrace the AI tipping pointSee patterns and insights in any amount of data Solve problems at scale with accuracyDemocratize access to ML and AI0102032023 Data and AI Trends Report19Embrace the AI tipping pointEmbrace the AI tipping pointAI-powered experiences are

43、 now embedded into everyday life.This ubiquity is creating demand for easier ways that more people can work with AI and ML.Trend 3Trend 3June YangVP,Cloud AI and Industry SolutionsGoogle Cloud2023 Data and AI Trends Report20Embrace the AI tipping pointWeve reached the AI tipping point.Whether people

44、 realize it or not,were already using applications powered by AI every day.Social media platforms,voice assistants,and driving services are easy examples.Organizations are adopting AI and ML tools and technologies because with them,they can pull out so much more information from the data they have a

45、nd solve real-world problems with scale and accuracy.Unification is the most important aspect to emerging AI adoption strategies.Even as early as a year ago,companies were thinking about and managing their data clouds and their AI clouds as separate entities.But as were seeing in other trends,this s

46、eparation or siloed strategy creates obstacles.Today,data scientists,analysts,developers,and other ML creators are all working together.And they all want a single interface where they can get their tools,their data,and their insights all through a single,unified portal.”“where they can get their too

47、ls,their data,and their insights all through a single,unified portal.”.data scientists,analysts,developers,and other ML creators.all wanta single interfaceTrend 3Overcoming the ML skills gap Organizations across industries that have made AI/ML more accessible to more staff are advancing the way they

48、 operate.For instance,retailers lean into AI/ML to:Serve personalized recommendations to their shoppers Ensure product availability by forecasting demand Give special attention to customers that need it with churn forecastingFinancial services and insurance companies use AI/ML to:Gain advanced fraud

49、 detection capabilities Classify and translate documents Analyze transactions and detect anomaliesTelecommunications organizations deploy AI/ML to:Automate their contact centers with virtual agents that mitigate common caller concerns Preserve live agents time for complicated or urgent matters Monit

50、or cell towers automatically Identify useful data trends and predictions80%of organizations say that having embedded support for AI/ML model execution makes them more likely to choose a particular data cloud platform.102023 Data and AI Trends Report21Embrace the AI tipping pointBecause most companie

51、s dont have the data science staff they need to meet their AI/ML goals,more organizations are empowering“citizen data scientists”to develop ML models using pre-trained models or low-code training methods.And 81 percent of organizations state that having more citizen data scientists would substantial

52、ly improve their ability to apply advanced analytics to more projects.9Trend 3Macys delivers tailor-made recommendations and search results to customers with machine learning.AmericasEuropeAsiaLufthansa cuts Co2 emissions by an estimated 7,400 tons per year with predictive intelligence.Jardine Resta

53、urant Group sees a 30 percent increase in average order value with AI-powered menu recommendations.Tips for AI/ML adoption Even if you know data science well,it doesnt mean you want to start everything from scratch.Use templates,models,and other ready-to-use assets that allow for customization and t

54、ake care of 80 percent of the work,so you can focus your efforts.Model tracing is critical to understand including when it was trained,who trained it,and where the data came from.You dont need to build a rocket ship,you just need to build a model that does a task better than what youre doing already

55、.Tackle small,quick-win projects.Using ML to improve search click-through rates by 3 or 4 percent may not sound like a sexy number but in reality,that little project could mean millions of dollars in additional revenue.Successful AI solutions build reliability and stability into the model at the out

56、set.2023 Data and AI Trends Report22Embrace the AI tipping point“One of the cool things were playing with is recommendations.How do I make sure you got the right sheets and right towels,if youre buying a comforter for your bed.With the power that were enabling Google Cloud,were going to help with th

57、ese recommendations.”Lauren Miller,Chief Information Officer,Macys2023 Data and AI Trends Report23Infuse insights everywhere04Infuse insights everywhere75%of organizations will demand new decision support features unavailable in their legacy BI software in the coming years.11Trend 42023 Data and AI

58、Trends Report24Infuse insights everywhereImprove decision makingFast development of new revenue streamsImprove customer acquisition and retention0102032023 Data and AI Trends Report25Infuse insights everywhereInfuse insights everywhereImproving decisions,customer services,product development,and rev

59、enue by rethinking BI and analytics strategies and applications.Trend 4increased their use of process automation1374%improved the quality of decisions across their organization1478%Over the past 18 months,as a result of investment in data,analytics,and AI/ML:73%improved delivery of actionable inform

60、ation to all users in the flow of work12Trend 4Kate WrightSenior Director,Product ManagementGoogle Cloud2023 Data and AI Trends Report26Infuse insights everywhereSpending on data and analytics technologies is forecasted to reachworldwide in 2026.15$200BDespite years of significant investment in data

61、 and analytics,BI has struggled to gain widespread adoption in organizations.One reason stems from a lack of trust in the reports and the tools themselves.Traditional reports often deliver inconsistent or inaccurate data because theyre created using stale data copies,siloed tools,and non-standard ca

62、lculations.Another reason for slow BI adoption is that the output is often a shared dashboard that provides general metrics rather than clear,actionable take-aways,tailored for specific users.In order to accelerate adoption,organizations are changing their expectations for BI,including traditional d

63、ashboard modalities.Theyre exploring different solutions to deliver context-rich data experiences that give users the information they need,when and where they need it.This pertains to all usersnot just the data savvy data analysts who know SQL.Organizations are equipping business decision-makers wi

64、th the tools they need to incorporate required insights into their everyday workflows.As organizations rethink BI and engage with multiple solutions,theyll need to make sure theyre drawing from consistent data metrics and definitions in real time to ensure one version of truth.Additionally,instead o

65、f measuring BIs ROI by the amount of times someone has logged into a dashboard,it will be important to measure the outcomes of improved decision making such as increased revenue,optimized supply chains,and innovative product development.”“Trend 4More than just KPIs2023 Data and AI Trends Report27Inf

66、use insights everywhereBI is on the move.Forward-looking organizations are leaving the traditional,dashboard-focused model behind in favor of adopting an action-focused BI paradigm where insights are served to more people in more environments to support more types of workflows than ever before.Organ

67、izations are using BI and analytics to identify underlying trends but also data anomalies and underlying business issues.These insights do not necessarily have to involve ML or AI.However,its important to note that 87 percent of organizations find it important for BI and analytics software to suppor

68、t the development and deployment of predictive models.16 In these use cases,BI and analytics feed data into models to deliver instant insights to users,even in dynamic,milliseconds-count environments such as digital ad bidding.Other use cases such as embedding BI into enterprise applications,which i

69、s critical for 87 percent of organizations,help reach more personas.This is important because 79 percent of organizations want to reach a broader internal audience and 66 percent want to reach more external users with built-in BI and analytics capabilities in their enterprise applications.17Embeddin

70、g analytics into customer-facing applications also improves service levels and creates new revenue streams.Organizations are creating deeply personalized omnichannel experiences with data,optimizing inventory and product placement decisions,and increasing the visibility and efficiency of their suppl

71、y chainsall supported by a modern business intelligence platform.79%of organizations want to reach a broader internal audience and 66%want to reach more external users with built-in BI and analytics capabilities in their enterprise applications.17Trend 4Tip:Build consistent trusted metrics via a sem

72、antic layerSemantic layers sit on top of your data,and control what data users can see.They also define the data and map relationships to related data.To reduce complexity,serve up consistent insights to all your users,and enable greater data exploration,create a consistent semantic layer for people

73、 to interact with,rather than just raw data.To improve efficiency,people only need to see the data thats relevant to them,and they need to know that its accurate and up to date.Flashpoint gives customers real-time insights into security threats with embedded analytics.AmericasEuropeAsiaAuto Trader s

74、atisfies data-hungry employees and customers with scalable,trusted,self-service data access.Mitsubishi Heavy Industries shares IoT data analysis across the organization to achieve better customer experiences and higher lifetime value.“The big advantage of Looker is its data modeling layer,LookML,whi

75、ch serves as a single source of truth for the whole company.Thats really important if you have a large team of analysts working across different business areas.”Edward Kent,Principal Developer,Data Engineering,Auto Trader2023 Data and AI Trends Report28Infuse insights everywhere2023 Data and AI Tren

76、ds Report29Get to know your unknown data05Get to know your unknown dataToday,77 percent of organizations are looking to improve their ability to classify data and enforce data security and privacy controls.18Trend 52023 Data and AI Trends Report30Get to know your unknown dataImprove productivity and

77、 collaboration Increase trust with your customersReduce risk of compliance breaches and fines0102032023 Data and AI Trends Report31Get to know your unknown dataGet to know your unknown dataOrganizations are looking to uncover and mitigate regulatory and compliance risks caused by their unknown data.

78、Trend 5Trend 5Anton ChuvakinSenior Staff Security ConsultantGoogle Cloud2023 Data and AI Trends Report32Get to know your unknown dataData is valuable.Its a big part of what makes companies competitive.But as companies amass volumes of structured and unstructured data from more channels supporting cu

79、stomers,partners,suppliers,and employees,many are unaware of the level of risk all this data brings.If you dont know what data you have,you cannot secure it.You also dont know what security risks you are incurring,or what security measures you need to take.If you create a table in a database that ha

80、s personally identifiable information(PII)like patient data,you should know what kind of data will be in it,how to secure it,and how to keep it compliant.But modern enterprises are collecting and copying large amounts of data,especially unstructured data,from many sources,and they are finding that i

81、ts impossible to manually find,scan,and classify every data set for risk.Unstructured data from chat applications or log files can cause significant headaches for organizations,especially if they unexpectedly contain sensitive data like PII.An example of this is customer support transcripts,because

82、you never know what information people will submit.When someone chats with customer support,they could type,“I didnt get my medications.Heres my name,the medications I need,and my social security number.”That sensitive PII data is now in one of your databases which may not be appropriately secured a

83、nd classified.“data from chat applications or log files can cause significant headaches for organizations,especially if they unexpectedly contain sensitive data like PII.”UnstructuredTrend 52023 Data and AI Trends Report33Get to know your unknown dataMake sure your data is discoverableGaining visibi

84、lity into all your data is the most critical first step in data risk management.This includes understanding all of your data ingestion pipelines and storage silos.Trend 52023 Data and AI Trends Report34Get to know your unknown dataClassify your data73%of organizations moved closer to having a common

85、 language around all data assets or artifacts2072%increased trust in data,information,and insight21Once you know where your data is,you need to classify all of it.Accuracy is key.Because theres often no way to do this manually,organizations are augmenting current skills and resources using machine l

86、earning and business automation tools.As we learned in trend four,90 percent of companies are also using BI and analytics to detect anomalies in data.19 Using anomaly detection in this way could flag any data types that dont conform with the purpose of a table or file store.Trend 52023 Data and AI T

87、rends Report35Get to know your unknown dataImplement consistent controlsOnce you have discoverable data thats also classified,you can implement automated controls that help reduce risk as data is stored and shared.For example,if you suspect that customers will be providing sensitive data like PII wh

88、en they interact with a customer service representative,you can configure an automated process that automatically takes steps like:Redacting the customers PII before the transaction information is stored in your system Storing all of the transaction data but tokenizing the PII if the transaction dat

89、a ever leaves the system where its stored Storing all of the transaction data but blocking it from being moved to certain states or regionsTrend 52023 Data and AI Trends Report36Get to know your unknown dataProactive risk management in actionRetailers are protecting personally identifiable informati

90、on and other sensitive data that unexpectedly show up customer support lines and in product reviews on websites.Manufacturing and logistics companies are meeting sovereignty demands and exercising control over where data resides,as well as understanding and mitigating risks from IoT data generated b

91、y sensors in factories and on vehicles.Financial Services and insurance organizations are running proactive risk reports and protecting PII and other sensitive customer data,including information exchanged on customer support calls.Trend 5Emerging changes in this trendBecause data security is such a

92、 complex issue,more companies are taking a collaborative approach.By 2025,growth in data marketplaces,data privacy regulations,and data sovereignty concerns will lead 60 percent of G2000 organizations to include chief data officers,along with chief information security officers,and chief legal offic

93、ers to join data risk management committees.22Ambra Health establishes an open source medical imaging data set that meets global security and privacy regulations to enhance global researchers collaboration and deep learning,and improve patient care.AmericasEuropeAsiaScotiabank moves its PII to the c

94、loud using a strategy that constrains access,and carefully and selectively allows reidentification by bank applications.Sunway Group classifies and protects sensitive data from over 10 sources to keep“green cities”running smoothly.2023 Data and AI Trends Report37Get to know your unknown data“At firs

95、t,we didnt want to place all of our data in the cloud for automated cleansing,structuring,and flow.However,as we navigated the process,we realized Cloud DLP and Cloud Key Management would help us navigate local country policies around data privacy.As we began to realize the potential of Google Cloud

96、,we put a larger portion of our data into the service and ran a larger number of integrations.”Amar Catic,Sales Strategy Manager,Swisscom66%of large enterprises will make major investments in data control plane technologies that can measure the risk inherent in data and reduce risk through security

97、and screening.23By 2027Make Data Work for You38Technologies to considerYou can choose from a vast landscape of technologies to incorporate these trends into your business strategies.Googles Data Cloud can help with your planning,as can many Google Cloud technology partners:Aiven:A fully managed,open

98、 source cloud data platform that helped logistics company Swift expand its delivery and order fulfillment services C3AI,Elastic,Plato Alto Networks:A broad ecosystem of specialized technology solutions that can help you go from data chaos to clarity CockroachDB:A distributed SQL database that helps

99、scheduling app Booksy build resilient architecture to keep up with global customer demands Confluent:Data streaming services used by Cargo Signal to optimize IoT sensor data pipelines and provide logistics enrichment services to all supply chain stakeholders Collibra:Unified governance and unified v

100、iews of data across multicloud storage Collibra,Confluent,Fivetran,Databricks,Informatica,Tableau:Embracing a Complete Cloud Data Platform on Google Cloud Databricks:Data lakehouse architecture and AI company that improves Reckitts marketing ROI with AI Datametica:Data migration tools used by a heal

101、thcare insurer to seamlessly migrate a critical data warehouse to Google Cloud Elastic:Observability solution that gives retailer Auchan France a clearer view of data without maintaining infrastructure,freeing time to focus on strategic analytics Fivetran:Save Nandos 80%of time by automating ELT&dat

102、a integration workflows MongoDB:An open source database used by Google Cloud and Forbes for data-driven insights NVIDIA:Accelerator-optimized solutions that help Cash App speed up a core ML workflow by 66 percent Qlik:A data integration platform that establishes real-time data replication between SA

103、P and BigQuery Quantiphi:Cloud-based machine learning services used by John Hopkins University BIOS Division to help brain injury patients SAP:ERP software provides critical data services in ATAs sustainable utilities model that support millions of customers Striim:Continuous,real-time data movement

104、 to Google Cloud ThoughtSpot:A search and AI-driven analytics platform for use by business people2023 Data and AI Trends Report39Ready to take your next steps?Weve talked a lot about the changing landscape of data and AI,and how organizations use both to:Break down data silos Take advantage of all t

105、hat an open data ecosystem has to offer Expand AI adoption by empowering citizen data scientists Become more insights-driven by rethinking BI and analytics Proactively manage data risksTo learn more about how Google Cloud can help you stay current and competitive,please contact us.Talk to an expertA

106、ppendix2023 Data and AI Trends Report40IDCs methodology for this researchIDC conducted a study,underwritten by Google,of over 800 global organizations to understand these three questions:What are the biggest challenges organizations are facing in using their data?What benefits are companies getting

107、today using data and AI cloud solutions?Where are companies going next with data and AI solutions?Future of Intelligence SurveyFuture Enterprise Resiliency and Spending SurveyIDC Global DataSphereIDC Business Intelligence and Analytics SurveyIDC Data Trust SurveyIDC Data as a Service Survey010402050

108、306This report also includes additional data points from other IDC syndicated studies and data products such as:2023 Data and AI Trends Report41Footnotes2023 Data and AI Trends Report421 Unified Data Cloud for Simplicity and In-telligence to Drive Better Business Out-comes,IDC Doc#US48822522,a White

109、 Paper sponsored by Google,March 2022.2 Revelations in IDCs Global DataSphere,2022,IDC#US49643822,September 2022.3 Sourcing and Utilizing External Data,2022:Data Buyer Profiles and Preferences,IDC Doc#US47715422,May,2022.4 Unified Data Cloud for Simplicity and Intelligence to Drive Better Business O

110、ut-comes,IDC Doc#US48822522,a White Paper sponsored by Google,March 2022.5 State of the Open-Source DBMS Market,Gartner6 Sourcing and Utilizing External Data,2022:Data Buyer Profiles and Preferences,IDC Doc#US47715422,May,2022.7 Sourcing and Utilizing External Data,2022:Data Buyer Profiles and Prefe

111、rences,IDC Doc#US47715422,May,2022.8 IDC FutureScape:Worldwide Artificial In-telligence 2020 Top 10 Predictions9 Self-Service Analytics in the Age of Ma-chine Learning,IDC Doc#US48733822,May 2022.10 Unified Data Cloud for Simplicity and Intelligence to Drive Better Business Out-comes,IDC Doc#US48822

112、522,a White Paper sponsored by Google,March 2022.11 Business Intelligence and Analytics Sur-vey,IDC,2022.12 Synthesizing Information for Enterprise Intelligence:Overcoming Information Overload,IDC Doc#US48464421,Decem-ber 2021.13 Synthesizing Information for Enterprise Intelligence:Overcoming Inform

113、ation Overload,IDC Doc#US48464421,Decem-ber 2021.14 Synthesizing Information for Enter-prise Intelligence:Overcoming Information Overload,IDC Doc#US48464421,Decem-ber 2021.15 Worldwide Big Data and Analytics Spending Guide,August 2022.16 Self-Service Analytics in the Age of Ma-chine Learning,IDC Doc

114、#US48733822,May 2022.17 Staying in the Flow with Embedded Ana-lytics,IDC Doc#US49375922,June 2022.18 How Much Do We Trust or Not Trust Data?:Key Findings from IDCs 2022 Data Trust Survey,IDC Doc#US46382820,Feb-ruary 2022.19 Self-Service Analytics in the Age of Ma-chine Learning,IDC Doc#US48733822,May 2022.20 Synthesizing Information for Enter-prise Intelligence:Overcoming Information Overload,IDC Doc#US48464421,Decem-ber 2021.21 Ibid.22 IDC FutureScape:Worldwide Data and Content Technologies 2023 Predictions,IDC Doc#US48733222,October 2022.23 Ibid.