AI as a Grantmaking Filter: Promise, Pitfalls, and the Human Judgment Gap

Imperial College London is piloting an AI-driven approach to spot climate research with commercialization potential, awarding small, flexible grants without IP strings. Proponents argue AI can scale triage and improve equity and cross-field insight, while critics warn of bias and lock-in to past successes. Public funders remain wary over confidentiality, and experts stress the need for rigorous testing; AI will likely augment, not replace, human judgment.

Key Points

• Imperial’s CSC used a tailored ChatGPT workflow to scan 10,000 UK research abstracts, shortlist 160, and—after human review—award three no-strings £35,000 grants aimed at commercialization steps.
• The program seeks to surface hidden, high-potential climate solutions and fund activities traditional grants often miss (e.g., industry engagement, market analysis) without taking equity or IP.
• Advocates say AI can scale triage, improve cross-disciplinary understanding, and potentially mitigate inequities by proactively reaching overlooked researchers.
• Concerns center on algorithmic bias and lock-in: evidence from VC suggests AI can favor look-alike bets, which may stifle novelty.
• Public funders are cautious due to confidentiality and bias risks—NIH banned AI in peer review and UKRI forbids reviewers’ use of generative AI—so more testing is needed before broad adoption.

Sentiment

The community is broadly skeptical of using AI as a grantmaking filter, with most commenters seeing significant risks around bias reinforcement, adversarial gaming, and the undermining of scientific judgment. However, there is meaningful support for limited AI augmentation — particularly as a pre-review self-assessment tool and for proactive researcher scouting — and a near-consensus that the existing grant process is already deeply flawed. The divide is less about whether AI has a role and more about how central that role should be.

In Agreement

• AI can serve as an effective large-scale triage tool to surface overlooked researchers and promising work that traditional processes miss
• The current grant review process is already so dysfunctional and formulaic that AI augmentation is unlikely to make it worse
• AI is useful as a pre-review tool for grant applicants to stress-test their own proposals before submission
• Proactive AI-driven outreach (talent scouting) is a genuinely innovative approach that could improve equity in funding

Opposed

• AI will inevitably create an arms race between AI-written proposals and AI reviewers, degrading the scientific legitimacy of the entire process
• AI functions as a conventional wisdom machine that would reinforce existing biases toward safe, incremental research and against breakthrough innovation
• LLMs can be fooled by their own confident-sounding output, rewarding unfounded boasting in grant applications rather than genuine scientific merit
• Using AI as a filter puts reviewers in an impossible position — they would need to review all rejected proposals to verify the AI's decisions, negating any efficiency gains
• Prompt injection and adversarial attacks on grant-reviewing AI represent a real and largely unsolvable vulnerability