top of page

AI in Biotech: Navigating Innovation, Capital, and Risk in a Rapidly Growing Market

Ailin Zhang

2025/11/04

AI in Biotech: Navigating Innovation, Capital, and Risk in a Rapidly Growing Market

Artificial intelligence (AI) is increasingly integrated into every step of the scientific pipeline from early-stage discovery to clinical solutions. With the global AI in biotech drug discovery market size projected to grow from $6.93 billion in 2025 to $16.52 billion by 2034, investments from both public and private sectors are rapidly scaling (1). Key funding areas include infrastructure, research and governance ($37.3 billion), data management and processing ($16.6 billion), and medical and healthcare applications ($11 billion), all of which directly support the advancement of the biotech sector (2). However, alongside this growth are critical concerns around data limitations, environmental impact, and the ethical risks of deploying imperfect AI systems in high-stakes medical contexts. This article explores the global AI investment surge in biotech, the key players driving it, and the pressing challenges that must be addressed to ensure its responsible and sustainable evolution.


The Global Race to invest in AI


Governments around the world are making strategic, long-term investments to build national AI ecosystems that support sectors like biotech. In North America, the United States announced a $500 billion investment into the Stargate initiative and Canada introduced a $2.4 billion AI infrastructure package (3-4). In Asia, China launched a $8.2 billion fund aimed at early investments in AI projects, while India pledged $1.25 billion and Saudi Arabia’s Project Transcendence included a massive $100 billion commitment (5-7). The European Union aims to invest 20 billion Euros to accelerate Europe’s AI projects (8). While France pledged to mobilize a total of 109 billion Euros via private investment toward the nation’s AI infrastructures (9). As part of the investment, the UAE would invest 30-50 billion Euros in France (10). 


Europe also leads in collaborative innovation. The MELLODDY project received $21 million in public funding and brought together 10 pharmaceutical companies and six partners to build an AI tool using federated learning (11). This approach allows companies to collaboratively train AI models on sensitive datasets without sharing proprietary data. MELLODDY’s success has inspired similar initiatives globally, including a new effort in South Korea to harness AI collaboration among domestic firms (12).


Meanwhile, the private sector has emerged as a dominant force in AI development, with global private AI investment surging by 44.5% in 2024 to a record $252.3 billion, much of it funneled into healthcare, life sciences, and biotechnology (2). Meanwhile the United States led in private AI investment with $109.1 billion, far outpacing China’s $9.3 billion and the UK’s $4.5 billion, reinforcing biotech as a central arena for global AI competition. Over 2,000 AI companies and 214 generative AI startups were newly funded, accelerating biotech innovation through corporate R&D, venture capital, and startup agility.


Major firms like Amgen, GSK, Roche, and Johnson & Johnson are ramping up AI programs in drug discovery, clinical trial optimization, and real-time public health monitoring (13-16). These efforts are increasingly supported by partnerships with startups. UK-based BenevolentAI integrates AI with wet-lab science and collaborates with AstraZeneca and Merck to identify novel drug targets (17). Hong Kong’s Insilico Medicine has developed its Pharma.AI platform to generate new drug candidates, including one currently in Phase 2 trials (18). France’s Owkin employs its agentic AI system, Owkin K, to uncover causal biological insights and advance precision medicine, working closely with major pharmaceutical players like Sanofi and BMS (19-20).


Challenges and Ethical Considerations


Data Limitations and Model Performance: The rapid advancement of AI in biotechnology brings with it several challenges related to data availability and model performance. Research from the Epoch AI team projects that the current stock of training data will be fully utilized between 2026 and 2032 (21). Overtraining and excessive reliance on synthetic data can degrade model performance and lead to model collapse (22). Supplementing real data with synthetic data rather than replacing it can help mitigate this degradation, though it does not necessarily improve model performance.


Energy Consumption and Environmental Impact: Despite significant improvements in energy efficiency, the total power draw for training cutting-edge AI systems continues to rise dramatically. Models like Google’s PaLM and Llama 3.1 405B, consume millions of watts of power, with energy demand doubling annually. This surge in power usage directly correlates with increasing carbon emissions: GPT-3’s training emitted around 588 tons of CO2, GPT-4 over 5,000 tons, and Llama 3.1 405B nearly 9,000 tons. This raises major sustainability concerns, especially as AI becomes integral to biomanufacturing and drug development (2).


Ethical Considerations: The ethical implications of AI inaccuracies are particularly critical due to the high stakes involved in healthcare decision-making and drug development. The Hughes Hallucination Evaluation Model (HHEM) leaderboard, developed by Vectara, shows that popular models, such as ChatGPT and Gemini, still hallucinate at a rate of ~1 to 3% (23). Furthermore, Google’s FACTS Grounding found Gemini 2.0 leading the board with 83.60% in factuality score (24). Inaccurate AI outputs in biotech could lead to flawed hypotheses, misidentified drug targets, or even compromised patient safety if clinical decisions are made based on incorrect data. As such, robust oversight, validation protocols, and transparency in model outputs are essential to safeguard patient outcomes and maintain scientific integrity.


Conclusion


The global race to invest in AI, spanning both public and private sectors, is transforming biotechnology at an unprecedented pace. From government-backed national AI initiatives to startup–pharma collaborations, capital is pouring into infrastructure, data systems, and medical applications, accelerating drug discovery and redefining clinical development. Yet, as AI becomes more deeply embedded in healthcare decision-making, the risks of flawed outputs, unsustainable energy consumption, and ethical lapses grow more pronounced. Navigating these challenges will require coordinated global efforts to promote data integrity, model transparency, and sustainable AI practices. If harnessed responsibly, AI in biotech holds the potential not just to revolutionize medicine, but to do so in a way that is accurate, ethical, and equitable for all.


References:


  1. Artificial Intelligence (AI) In Drug Discovery Market Size, Report 2030. Precedence Research. https://www.precedenceresearch.com/artificial-intelligence-in-drug-discovery-market

  2. Artificial Intelligence Index Report 2025. Stanford University. https://hai.stanford.edu/assets/files/hai_ai_index_report_2025.pdf

  3. Announcing The Stargate Project. Open AI. https://openai.com/index/announcing-the-stargate-project

  4. Securing Canada’s AI advantage. Prime Minister of Canada. https://www.pm.gc.ca/en/news/news-releases/2024/04/07/securing-canadas-ai

  5. Cao, A. (2025, April 11). New AI fund in China to pour US$8 billion into early-stage projects. South China Morning Post. https://www.scmp.com/tech/policy/article/3306047/new-ai-fund-china-pour-us8-billion-early-stage-projects

  6. ‌Ghosh, S. (2024, March 17). India’s US$1.25 billion push to power AI. Nature India. https://doi.org/10.1038/d44151-024-00035-5

  7. Vella, H. (2024, November 20). Saudi Arabia Launches $100B Initiative to Develop AI Ecosystem [Review of Saudi Arabia Launches $100B Initiative to Develop AI Ecosystem]. AI Business. https://aibusiness.com/responsible-ai/saudi-arabia-launches-100b-initiative-to-develop-ai-ecosystem

  8. EU launches InvestAI initiative to mobilise €200 billion of investment in artificial intelligence. European Commission. https://digital-strategy.ec.europa.eu/en/news/eu-launches-investai-initiative-mobilise-eu200-billion-investment-artificial-intelligence

  9. Browne, R. (2025, February 10). France unveils 109-billion-euro AI investment as Europe looks to keep up with U.S. CNBC. https://www.cnbc.com/2025/02/10/frances-answer-to-stargate-macron-announces-ai-investment.html

  10. NEWS WIRE. (2025, February 7). UAE to invest billions in France AI data centre. France 24. https://www.france24.com/en/europe/20250207-uae-to-invest-up-to-%E2%82%AC50-billion-in-massive-ai-data-centre-in-france

  11. Machine learning ledger orchestration for drug discovery. IHI Innovative Health Initiative. https://www.ihi.europa.eu/projects-results/project-factsheets/melloddy

  12. https://kmelloddy.org/english/aboutus

  13. AI in Research & Development. Amgen. https://www.amgen.com/science/research-and-development-strategy/ai-in-research-and-development

  14. “We’ve seen an explosion in computing power”: Using AI, machine learning and data to unlock the mysteries of disease. GSK. https://www.gsk.com/en-gb/behind-the-science-magazine/ai-ml-data-computing-power/

  15. Harnessing the power of AI. Roche. https://www.roche.com/stories/harnessing-the-power-of-ai

  16. Welch, A. (2023, September 14). Artificial Intelligence is Helping Revolutionize Healthcare As We Know It. Johnson & Johnson. https://www.jnj.com/innovation/artificial-intelligence-in-healthcare

  17. BenevolentAI and AstraZeneca collaboration yields continued success as further novel target progressed into portfolio. BenevolentAI. https://www.benevolent.com/news-and-media/press-releases-and-in-media/benevolentai-and-astrazeneca-collaboration-yields-continued-success-further-novel-target-progressed-portfolio/

  18. First Generative AI Drug Begins Phase II Trials with Patients. Insilico Medicine. https://insilico.com/blog/first_phase2

  19. Sanofi invests $180 million equity in Owkin’s artificial intelligence and federated learning to advance oncology pipeline. Sanofi. https://www.sanofi.com/en/media-room/press-releases/2021/2021-11-18-06-30-00-2336966

  20. Macdonald, G. (2022, June 9). Bristol Myers pays $80M to AI firm Owkin as part of cardiovascular trial accord. Fierce Biotech. https://www.fiercebiotech.com/cro/bristol-myers-pays-80m-ai-firm-owkin-part-cardiovascular-trial-accord

  21. Villalobos, P. (2024, June 6). Will We Run Out of Data? Limits of LLM Scaling Based on Human-Generated Data. Epoch AI. https://epoch.ai/blog/will-we-run-out-of-data-limits-of-llm-scaling-based-on-human-generated-data

  22. Shumailov, I., Shumaylov, Z., Zhao, Y. et al. AI models collapse when trained on recursively generated data. Nature 631, 755–759 (2024). https://doi.org/10.1038/s41586-024-07566-y

  23. HHEM Leaderboard - a Hugging Face Space by vectara. Huggingface.co. https://huggingface.co/spaces/vectara/leaderboard

  24. FACTS Grounding: A new benchmark for evaluating the factuality of large language models. Google DeepMind. https://deepmind.google/discover/blog/facts-grounding-a-new-benchmark-for-evaluating-the-factuality-of-large-language-models/

bottom of page