Introduction

Artificial intelligence (AI) has emerged as a genuine general‑purpose technology—comparable in impact to electrification or the internet. It is reshaping competitive dynamics, supply chains, consumer expectations and even geopolitics. Global AI spending (software, services and hardware) is on track to reach hundreds of billions of dollars annually this decade, with some forecasts pointing to AI adding up to US$15.7 trillion to global GDP by 2030.^[3]

For business leaders, AI is no longer a side experiment. It sits at the center of strategy: how to drive productivity, differentiate customer experience, build defensible capabilities and manage new classes of risk. At the same time, governments race to define national AI strategies, build domestic AI capacity and design regulatory frameworks that both enable and constrain AI’s development.

This report provides a comprehensive, board‑level view of the global AI ecosystem in 2025. It covers:

• Market size, investment and adoption trends.
• Regional ecosystems and national AI strategies.
• Core technologies and emerging capabilities (including generative AI).
• Leading companies, startups and research institutions.
• Policy, regulation, ethics and responsible AI.
• Impacts on labor, education and economic development.
• Sector‑by‑sector applications and strategic implications.

Global Market Size & Investment Trends

The AI economy has moved from hype to scale. Spending, investment and corporate commitment all point in the same direction: AI is now a multi‑hundred‑billion‑dollar market on its way to the trillion‑dollar range.

Estimates vary, but most major analyst houses now see the global AI market (software, services and supporting hardware) in the US$250–300 billion range in 2025, with compound annual growth rates around 25–30 percent out to 2030.^[2] Many forecasts expect total AI‑related revenues to exceed US$1 trillion early in the 2030s. This is driven not only by core AI platforms, but by AI embedded in virtually every enterprise application and device.

Investment has scaled even faster than revenue. According to the Stanford AI Index, total global corporate investment in AI—including internal spending, acquisitions and venture funding—reached roughly US$250 billion in 2024, more than 13‑times higher than a decade earlier.^[1] Private venture funding in AI rebounded sharply after 2022’s tech downturn, with generative AI startups capturing a disproportionate share of capital.

The geographic concentration of AI capital remains striking. The United States accounts for well over 40 percent of private AI investment; in recent years US AI funding has exceeded that of China, the European Union and the United Kingdom combined.^[1] China remains the clear #2 in absolute AI funding, while Europe, the UK, Canada, Israel, Singapore and others are building vibrant but comparatively smaller ecosystems.

Meanwhile, governments are treating AI as strategic infrastructure. The United States, China, the European Union, India, Saudi Arabia, the United Arab Emirates and others have announced multi‑billion‑dollar AI programs, spanning basic research, cloud and compute infrastructure, semiconductor capacity, national data platforms and AI skills initiatives.^[1] The result is an unprecedented public‑private investment flywheel around AI.

AI Adoption & Global Diffusion

Adoption has accelerated sharply over the last 24 months—especially with the arrival of accessible generative AI tools. Recent global surveys suggest that roughly three‑quarters of organizations now use AI in at least one business function, up from roughly half in 2022–23.^[2] Generative AI alone moved from early experimentation to mainstream: within a year of ChatGPT’s launch, a majority of large firms reported pilots or production use of generative AI.

Adoption is broad‑based. Every major industry now reports material AI deployments: financial services, healthcare, manufacturing, logistics, retail, media, energy, agriculture and the public sector all feature serious AI programs. The intensity of adoption varies—tech, telecoms and financial firms lead—but no sector is untouched.

Regional adoption patterns are converging. North American firms retain a lead in breadth and depth of AI use, but the fastest year‑on‑year adoption jumps are now seen in China and Europe.^[1] Public attitudes toward AI also show a nuanced picture: optimism about AI’s benefits is particularly high in many Asian, Middle Eastern and Latin American countries, while attitudes in North America and parts of Europe are more cautious but trending slightly more positive over time.

Regional AI Ecosystems & Strategies

AI is global, but the ecosystem looks very different across regions. Policy choices, industrial structure, talent pools and capital markets shape distinctive strengths and weaknesses.

North America

North America—particularly the United States—remains the epicenter of AI innovation. The region hosts the largest concentration of AI talent, capital and compute, and is home to leading AI platform and cloud providers including Alphabet/Google, Microsoft, Amazon Web Services, Meta, OpenAI, Anthropic and NVIDIA.

The United States has historically taken a light‑touch, innovation‑first regulatory stance. More recently, the Biden administration’s Executive Order on Safe, Secure and Trustworthy AI and voluntary commitments from major AI developers introduced more structured expectations around safety testing, security, watermarking and transparency for frontier models.^[6] Canada, meanwhile, punched above its weight by pioneering a national AI strategy in 2017 and nurturing globally recognized research hubs in Montreal, Toronto and Edmonton.

Europe

Europe combines strong research excellence with emerging industrial capabilities and world‑leading regulation. Universities such as Oxford, Cambridge, ETH Zurich and EPFL, and labs like Google DeepMind, make Europe a powerhouse in AI research, including reinforcement learning, robotics and AI ethics.

On policy, the European Union is setting the global benchmark with the EU AI Act—a risk‑based horizontal regulation that imposes stringent requirements on “high‑risk” AI systems, sets transparency obligations and prohibits certain harmful uses.^[7] Member states have also adopted national AI strategies, often backed by multi‑billion‑euro investments. The UK, now outside the EU, is positioning itself as a hub for AI safety research and convening power, exemplified by the 2023 AI Safety Summit at Bletchley Park.^[9]

Asia‑Pacific

Asia‑Pacific is a second major engine of AI development. China is the standout: it leads the world in AI publication volume and AI‑related patents, and is home to powerful AI platforms from Baidu, Alibaba, Tencent and Huawei. Its 2017 New Generation AI Development Plan sets an explicit goal to lead the world in AI by 2030, supported by heavy R&D and infrastructure spending.

Japan and South Korea leverage deep industrial strengths in robotics, electronics and automotive to build AI capabilities into manufacturing, consumer devices and mobility. Sony, Samsung and LG are prominent regional corporate players. India, meanwhile, is emerging as a global AI services and solutions hub, building on its IT and business‑process strengths and a large English‑speaking, technically trained workforce.

Middle East

The Middle East is using AI as a lever for economic diversification. The United Arab Emirates created the world’s first Ministry of Artificial Intelligence and a dedicated AI university (MBZUAI), while Saudi Arabia’s National Strategy for Data & AI aims to make the kingdom a top‑tier AI economy by 2030. Both countries are investing tens of billions of dollars in AI infrastructure, research and human capital, and attracting leading AI companies to set up regional hubs.

Africa

Africa’s AI ecosystem is still early but growing quickly. Countries including Kenya, South Africa, Rwanda, Ghana and Nigeria have published AI strategies or roadmaps. Global technology firms have invested in African AI research centers—such as Google’s AI center in Accra and Microsoft’s Africa Development Center—focused on local challenges in agriculture, health and financial inclusion. Key constraints remain: digital infrastructure gaps, data availability, and the need for scaled AI education and training.^[1]

Latin America

Latin America is moving from experimentation to structured AI strategies. Brazil, Mexico, Chile, Colombia and Argentina have all launched national AI plans, often emphasizing AI for public‑sector modernization, agriculture, natural‑resource management and fintech. Startup ecosystems in São Paulo, Mexico City and Santiago are giving rise to AI‑enabled fintech, logistics and health platforms. Multilateral institutions like the Inter‑American Development Bank support responsible AI deployment and capacity building across the region.

Core AI Subfields & Technologies

Today’s AI landscape rests on several interlocking subfields. Understanding their roles helps leaders see where value is being created and where disruption may emerge.

Machine Learning & Deep Learning

Machine learning (ML) is the core engine of modern AI: systems that learn patterns from data instead of following fixed rules. Deep learning—a subset using multi‑layer neural networks—has driven most of the breakthroughs in vision, speech, language and recommendation systems over the last decade. As models have scaled and compute has become cheaper, deep learning systems now match or exceed human performance on an expanding set of narrow tasks, from image recognition to complex language benchmarks.^[1]

Generative AI & Foundation Models

Generative AI refers to models that can create new content—text, images, code, audio, video—based on patterns learned from very large datasets. Foundation models such as GPT‑4, Llama 2 and Claude are trained on trillions of tokens and can be adapted (fine‑tuned) for diverse downstream tasks. These models have unlocked powerful new capabilities: fluent conversation, rapid code generation, content creation, summarization and more, with generative AI adoption soaring across enterprise functions.^[2]

Computer Vision

Computer vision enables machines to interpret images and video. Deep convolutional networks and vision transformers now power applications from facial recognition and medical imaging to autonomous vehicles and quality inspection in factories. Regulators have begun to scrutinize certain high‑risk uses—such as real‑time biometric surveillance in public spaces—due to privacy and civil‑rights concerns, but the underlying technology continues to advance rapidly in accuracy, robustness and efficiency.

Natural Language Processing (NLP)

NLP allows machines to understand, generate and interact in human languages. Transformer‑based language models have dramatically improved machine translation, search understanding, conversational agents, summarization and information extraction. Large language models (LLMs) are increasingly being combined with tools, databases and business systems to create powerful “AI copilots” for coding, marketing, sales, legal and customer service work.

Robotics & Autonomous Systems

Robotics integrates AI perception, planning and control into physical systems—industrial arms, mobile robots, drones and autonomous vehicles. Traditional industrial robots are evolving into more flexible, collaborative systems (“cobots”) that can operate safely around humans. Self‑driving vehicle pilots from companies such as Waymo, Cruise, Tesla and Baidu Apollo illustrate what is technically possible, even if full Level‑5 autonomy remains a work in progress.

Reinforcement Learning

Reinforcement learning (RL) trains agents to make sequences of decisions through trial and error, guided by rewards. RL has delivered headline‑grabbing achievements in games—such as DeepMind’s AlphaGo and AlphaZero—and is increasingly applied to industrial control, recommendation systems, robotics and resource optimization (for example, data‑center energy efficiency). In combination with deep learning, RL supports more autonomous and adaptive AI systems.

Emerging AI Technologies & Trends

Several cross‑cutting trends are shaping the trajectory of AI and opening new strategic opportunities.

Edge AI & TinyML

Edge AI moves computation from centralized clouds to devices—smartphones, sensors, vehicles, industrial equipment—reducing latency, bandwidth usage and privacy risks. Specialized chips from NVIDIA, Qualcomm, Google Coral and others, coupled with model compression techniques and TinyML, enable increasingly sophisticated inference directly on devices. Edge AI is critical for autonomous vehicles, industrial IoT, AR/VR, smart cities and privacy‑sensitive applications.

AI Hardware & Efficiency

The AI boom is also a hardware story. GPUs from NVIDIA dominate training workloads, while TPUs from Google and dedicated accelerators from Intel, AMD and multiple startups compete in both cloud and edge markets. At the same time, algorithmic and systems advances have dramatically improved cost‑performance for both training and inference, making state‑of‑the‑art models increasingly affordable to deploy at scale.^[1]

Autonomous Agents & Multi‑Step Automation

Building on LLMs and reinforcement learning, developers are experimenting with AI “agents” that can decompose high‑level goals into tasks, call tools and APIs, interact with other agents and iterate toward outcomes. While still early and brittle, these agent frameworks hint at a shift from one‑shot AI responses to multi‑step, semi‑autonomous workflows—especially in software operations, data analysis, research, and business process automation.

Multi‑Modal AI

Multi‑modal models process and relate different data types—text, images, audio, video, code and even sensor data. GPT‑4 and similar systems that accept both text and images are early examples. Over time, multi‑modal AI will power richer digital assistants that can “see” your screen, understand documents, listen to meetings and act across channels, and will be central to advanced robotics and AR/VR.

Quantum & Next‑Generation AI

Quantum computing’s intersection with AI remains largely experimental, with research exploring quantum‑accelerated optimization and machine‑learning algorithms. While not directly impactful on 2025 business planning yet, it is a frontier to watch for the 2030s as hardware matures.

Leading Companies, Startups & Research Labs

AI progress is driven by a dense network of large technology platforms, nimble startups and world‑class research institutions working in complex collaboration and competition.

Technology Giants

A small set of technology giants account for a large share of frontier AI development and deployment:

• Google / Google DeepMind – inventors of the Transformer architecture, creators of AlphaGo and AlphaFold, leaders in search, cloud, advertising and consumer AI.
• Microsoft – deep partnership with OpenAI, infusing GPT‑class models across Microsoft 365 Copilot, Bing and Azure.
• OpenAI – creator of ChatGPT, GPT‑4 and DALL·E, and a central catalyst of the generative AI wave.^[11]
• Meta – open‑sourcing powerful LLMs (Llama family), advancing recommender systems and multi‑modal AI.
• Amazon Web Services – major provider of AI infrastructure and platforms, with extensive use of AI across retail, logistics and devices (Alexa).
• NVIDIA – dominant supplier of GPUs and AI‑optimized hardware, and a leading provider of AI software frameworks.
• IBM – enterprise AI (Watsonx), AI governance tools and research in trustworthy AI.
• Baidu, Alibaba, Tencent, Huawei – core pillars of China’s AI industry, spanning cloud, search, social media, fintech, telecoms and hardware.

Startups & Scale‑Ups

Below the platform layer, thousands of startups are building AI‑native products and vertical solutions. Notable examples include:

• Foundation & tooling – Anthropic, Cohere, Mistral AI, Hugging Face, and Databricks.
• Robotics & autonomy – Waymo, Cruise, Nuro, Boston Dynamics, and warehouse automation firms such as GreyOrange.
• Healthcare & life sciences – Insilico Medicine, PathAI, Tempus.
• Enterprise automation – UiPath, C3.ai, Palantir, Scale AI.

Research Labs & Universities

Academic institutions remain critical for foundational AI research and talent development. Leading universities include Stanford, MIT, Carnegie Mellon, UC Berkeley, the University of Toronto and the Mila/Vector/Amii institutes in Canada; Oxford, Cambridge, ETH Zurich and EPFL in Europe; and Tsinghua, Peking and the Chinese Academy of Sciences in China.^[1] Non‑profit labs such as the Allen Institute for AI and industry labs like Meta AI, Microsoft Research and IBM Research are also central contributors.

Global AI Strategies, Policy & Ethics

AI has become a strategic, regulatory and ethical priority for governments and institutions worldwide. More than 60 countries now have national AI strategies, and the number of AI‑related laws and regulations has surged in the last five years.^[1]

National AI Strategies

Most national AI strategies aim to:

• Boost AI R&D and innovation (through grants, tax incentives, public‑private partnerships).
• Develop AI talent pipelines and retraining programs.
• Strengthen digital and compute infrastructure.
• Support AI adoption in key sectors (health, manufacturing, agriculture, public administration).
• Define broad principles for trustworthy AI.

The United States, European Union and China each combine industrial policy (funding, semiconductor support, cloud infrastructure) with security and export‑control considerations. Middle‑income and emerging economies tend to emphasize AI for development, inclusion and public‑sector modernization.

Regulation & Governance

The global regulatory landscape is coalescing around several anchor frameworks:

• OECD AI Principles – high‑level guidelines for trustworthy, human‑centric AI adopted by dozens of countries.^[8]
• UNESCO Recommendation on the Ethics of AI – a global standard covering human rights, fairness, transparency and sustainability.^[10]
• EU AI Act – the first comprehensive, binding horizontal AI regulation, using a risk‑based approach with strict obligations on high‑risk systems.^[7]
• US Executive Order on AI – introduces safety testing, transparency and equity expectations for private and public sector AI, alongside sectoral guidance.^[6]
• China’s AI regulations – including rules on recommendation algorithms, deep synthesis and generative AI, emphasizing content controls, security reviews and alignment with national priorities.

Responsible AI & Ethics

Responsible AI programs typically focus on four themes:

• Fairness – detecting and mitigating bias in training data and models.
• Transparency & explainability – explaining model behavior to users and regulators.
• Safety & robustness – preventing harmful outputs, adversarial attacks and failures in high‑stakes applications.
• Privacy – protecting personal data via minimization, anonymization and privacy‑preserving techniques.

Leading organizations now maintain AI ethics guidelines, internal review boards and technical toolkits for bias testing, model documentation and red‑teaming. However, independent surveys show that only a minority of companies have fully mature responsible‑AI governance embedded into product and risk‑management processes.^[1]

Impact on Workforce, Education & Economy

AI will reshape work, skills and economic structure. The central question is not whether AI will change jobs, but how quickly, for whom, and with what net effect on productivity and inclusion.

The Future of Work

Estimates differ, but multiple studies suggest that a significant share of current work activities could be partially automated by AI—especially in routine cognitive and data‑processing tasks. One analysis by Goldman Sachs suggested that generative AI could expose the equivalent of 300 million full‑time jobs globally to automation, while also creating new roles and increasing demand in AI‑complementary occupations.^[4]

Importantly, real‑world experiments show that AI often augments rather than simply replaces workers. Studies of customer‑support agents and software developers using AI copilots show double‑digit productivity gains, with the largest improvements for less experienced workers—narrowing performance gaps and potentially enabling faster up‑skilling.^[1]

Education & Skills

Education systems are under pressure to adapt on two fronts: teaching about AI and teaching with AI. Many countries have added computer‑science and basic AI concepts to K‑12 curricula, and universities worldwide offer degrees in data science and machine learning. At the same time, AI tutors, adaptive learning platforms and language‑learning apps are spreading rapidly, raising both opportunity (personalized instruction) and concern (academic integrity, over‑reliance).

For adult workers, continuous learning becomes essential. Corporate reskilling programs, online learning platforms such as Coursera, edX and Udacity, and government‑funded training initiatives aim to equip millions with AI and data‑literacy skills.^[5]

Economic Development & Inequality

AI could be a powerful engine of growth: PwC estimates AI could add up to US$15.7 trillion to global GDP by 2030 through productivity gains and product innovations.^[3] However, gains are unlikely to be evenly distributed. Advanced economies and firms with strong digital foundations and access to capital are best positioned to capture value, while those slower to adopt risk widening productivity gaps.

Within countries, AI may increase wage dispersion if high‑skill, AI‑complementary roles see wage premiums while routine roles stagnate or shrink. Thoughtful policy—around taxation, social protection, education and labor regulation—will be needed to ensure that AI‑driven growth is inclusive.

Industry Applications of AI

AI is reshaping competitive dynamics across sectors. Below we highlight impacts in five major industries that matter for global business leaders.

Healthcare & Life Sciences

In healthcare, AI supports diagnostics (especially medical imaging), triage, workflow automation and drug discovery. Deep‑learning models now match or exceed specialist performance in identifying certain pathologies on images, and regulators have approved hundreds of AI‑enabled devices and algorithms.^[1]

In drug discovery, AI platforms from companies like DeepMind and Insilico Medicine are compressing timelines from target identification to pre‑clinical candidates.^[12] AI‑enabled virtual assistants, symptom checkers and remote‑monitoring tools extend care capacity, particularly in primary care and chronic‑disease management.

Financial Services

Financial institutions use AI for fraud detection, anti‑money‑laundering, credit scoring, market surveillance, algorithmic trading, customer service and personalization. AI‑driven risk models detect anomalous transactions in real time; robo‑advisors allocate client portfolios; and NLP tools help comply with complex regulatory regimes by mining regulatory texts and internal communications.

At the same time, regulators are closely examining AI use for explainability, fairness and stability—especially for credit and underwriting decisions and algorithmic trading.

Manufacturing & Industry 4.0

Manufacturing is a major beneficiary of AI through predictive maintenance, process optimization, computer‑vision quality control and flexible automation. Predictive‑maintenance models reduce unplanned downtime; computer‑vision systems catch defects earlier; and AI‑guided robots and cobots improve throughput and safety. AI‑driven demand forecasting and production planning help smooth supply chains and reduce inventory.

Logistics, Mobility & Supply Chain

Logistics providers use AI for routing, network optimization, dynamic pricing and warehouse automation. Routing optimization at companies such as UPS, DHL and FedEx has cut fuel consumption and improved on‑time delivery. In warehouses, mobile robots orchestrated by AI work alongside humans to move and pick goods.

In mobility, AI powers advanced driver‑assistance systems (ADAS), fleet‑management platforms and early robotaxi services. Over time, as autonomy matures and regulations evolve, AI could transform freight, transit and last‑mile delivery economics.

Marketing, Sales & Retail

Marketing and retail have been early adopters of AI for personalization and optimization. Recommendation systems from platforms like Netflix, Spotify and Amazon set consumer expectations for tailored experiences. Programmatic advertising relies on AI for audience targeting and real‑time bidding. Generative AI tools now write copy, generate imagery, segment audiences and power conversational commerce.

Retailers use AI for demand forecasting, assortment optimization, dynamic pricing and store‑layout analytics. Customer‑service chatbots reduce call‑center volumes and improve response times when deployed carefully and transparently.

Breakthroughs, Open‑Source & the AI Community

Recent Breakthrough Systems

Several high‑profile AI systems have shaped both technical progress and public perception:

• GPT‑4 and large language models – dramatically improved performance on language, reasoning and multi‑modal tasks, passing professional exams and enabling conversational AI at scale.^[11]
• AlphaFold – accurately predicting protein structures at scale, unlocking new possibilities in biology and drug discovery.^[12]
• Text‑to‑image and multi‑modal models – systems like DALL·E 2, Midjourney and Stable Diffusion, which democratized high‑quality image generation and sparked new debates around creativity and IP.

The Open‑Source AI Movement

Open‑source AI has become a powerful counterweight to closed, proprietary models. Community‑driven models like BLOOM, Stable Diffusion and the Llama family, together with platforms such as Hugging Face, give developers and enterprises broad access to cutting‑edge models and tools. The performance gap between top open and closed models has narrowed significantly on many benchmarks.^[1]

This open ecosystem accelerates innovation, supports localization (for languages and domains) and raises important questions about governance, licensing and security for powerful models that anyone can download and run.

Community, Conferences & Research Scale

The AI research community has exploded in size. Leading conferences such as NeurIPS, ICML, ICLR and CVPR receive tens of thousands of paper submissions annually, and top venues now include dedicated tracks for ethics, policy and societal impacts. Collaboration across academia and industry is intense, with many landmark models emerging from industry labs but fundamental advances and critical scrutiny coming from universities and non‑profits.

Conclusion

The global AI ecosystem in 2025 is a complex, fast‑moving system of technologies, markets, institutions and policies. AI is already delivering tangible productivity and innovation benefits across industries, and its economic contribution is poised to grow substantially in the coming decade.^[3]

For business leaders, this implies several imperatives:

• Build a clear AI strategy – aligned to business goals, grounded in realistic use‑cases and supported by data, platform and talent investments.
• Invest in people and change – up‑skill the workforce, redesign workflows and embed AI into day‑to‑day decision‑making, not just standalone pilots.
• Take responsible AI seriously – treat ethics, safety, security and governance as core to AI success, not an afterthought.
• Monitor the policy landscape – understand emerging regulations across jurisdictions and build compliance and reporting capabilities early.

AI is not a single technology or project; it is a new foundation for how organizations operate and create value. Those that move thoughtfully but decisively—combining ambition with responsibility—will shape the next decade of global business.

Sources, References & Additional Reading

1. Stanford Institute for Human‑Centered Artificial Intelligence (HAI), AI Index Report 2024. https://aiindex.stanford.edu/report/
2. McKinsey & Company, The State of AI in 2024. https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai-in-2024
3. PwC, Sizing the prize: What’s the real value of AI for your business and how can you capitalise? (2017). https://www.pwc.com/gx/en/issues/analytics/assets/pwc-ai-analysis-sizing-the-prize-report.pdf
4. Goldman Sachs Global Investment Research, The Potentially Large Effects of Artificial Intelligence on Economic Growth (2023). https://www.goldmansachs.com/insights/pages/generative-ai-could-raise-global-gdp-by-7-percent.html
5. World Economic Forum, The Future of Jobs Report 2023. https://www.weforum.org/reports/the-future-of-jobs-report-2023
6. The White House, Executive Order on Safe, Secure, and Trustworthy Artificial Intelligence (2023). https://www.whitehouse.gov/briefing-room/presidential-actions/2023/10/30/executive-order-on-safe-secure-and-trustworthy-artificial-intelligence/
7. European Commission, Proposal for a Regulation of the European Parliament and of the Council Laying Down Harmonised Rules on Artificial Intelligence (AI Act). https://digital-strategy.ec.europa.eu/en/policies/european-approach-artificial-intelligence
8. OECD, OECD Principles on Artificial Intelligence (2019). https://oecd.ai/en/ai-principles
9. UK Government, Bletchley Declaration by Countries Attending the AI Safety Summit (2023). https://www.gov.uk/government/publications/ai-safety-summit-2023-the-bletchley-declaration
10. UNESCO, Recommendation on the Ethics of Artificial Intelligence (2021). https://www.unesco.org/en/artificial-intelligence/recommendation-ethics
11. OpenAI, GPT‑4 Technical Report (2023). https://openai.com/research/gpt-4
12. DeepMind, AlphaFold: Using AI for Scientific Discovery. https://deepmind.google/technologies/alphafold/