Scientists should use AI as a tool, not an oracle
This commentary critiques the feedback loop between AI hype and scientific research, arguing that scientists who treat AI systems as oracles rather than tools produce flawed research that in turn amplifies further hype. The piece examines how uncritical adoption of AI in scientific workflows can compromise research integrity. It calls for a more epistemically disciplined approach to AI use in science.
Related guides (2)
Related events (8)
Could AI Slow Science? Confronting the Production-Progress Paradox
A commentary piece from AI Snake Oil explores the potential paradox whereby AI tools increase scientific output volume while simultaneously slowing genuine scientific progress. The piece examines how AI-assisted research production may prioritize quantity over quality, potentially crowding out deeper, slower-moving inquiry. This raises structural concerns about how AI integration into research workflows could reshape the incentive landscape of science.
Giving your AI a Job Interview
This commentary piece argues that as AI-generated advice becomes more consequential, users need systematic methods to evaluate AI reliability and quality—analogous to a job interview process. The author proposes frameworks for assessing AI outputs before trusting them for important decisions. The piece addresses the practical challenge of calibrating trust in AI systems across different use cases.
On Working with Wizards
A commentary piece from One Useful Thing exploring the metaphor of AI systems as 'wizards' and the challenge of working with them on the 'jagged frontier' of capabilities. The piece likely addresses how users can effectively verify and leverage AI outputs given the uneven and unpredictable nature of current model capabilities. As a tier-2 commentary source, it offers practitioner-level perspective on human-AI collaboration patterns.
An Opinionated Guide to Using AI Right Now
A tier-2 commentary piece from One Useful Thing offering opinionated guidance on which AI tools to use in late 2025. The piece likely surveys the current landscape of frontier models and recommends specific tools for specific tasks. As a practitioner-facing guide, it reflects the state of the AI tooling ecosystem as perceived by an influential commentator.
AI existential risk probabilities are too unreliable to inform policy
This commentary argues that numerical probability estimates for AI existential risk are epistemically unreliable and should not be used as a basis for policy decisions. The piece critiques the practice of assigning precise figures to speculative scenarios, characterizing it as pseudo-quantification that lends false credibility to uncertain claims. The author contends that such estimates are laundered speculation rather than grounded forecasting.
Your AI Use Is Breaking My Brain
Simon Willison comments on the phenomenon of AI-generated or AI-assisted content degrading the quality of online discourse and information environments. The piece reflects on how widespread AI use is affecting the experience of consuming internet content. This is a commentary piece from a prominent developer/blogger on the social and epistemic effects of AI proliferation.
Against "Brain Damage": AI's Effect on Human Thinking
This commentary from One Useful Thing examines whether AI use helps or harms human cognitive capabilities. The piece engages with the ongoing debate about whether reliance on AI tools degrades or augments human thinking. It likely addresses concerns about cognitive offloading and the conditions under which AI assistance is beneficial versus detrimental.
Making AI Work: Leadership, Lab, and Crowd
This commentary from One Useful Thing proposes a framework for organizational AI adoption centered on three elements: leadership commitment, structured experimentation (lab), and distributed employee engagement (crowd). The piece offers practical guidance for companies navigating AI integration. As a tier-2 commentary source, it reflects practitioner thinking on enterprise AI deployment patterns rather than reporting new technical developments.