The trouble with these calculations is that they mire us in epistemically tricky terrain. I’m bothered by how quickly the discussions of AI become utopian or apocalyptic.
...
It also forces thinking to be obsessively short term. People start losing interest in problems of the next five or ten years, because superintelligence will have already changed everything. The big political and technological questions we need to discuss are only those that matter to the speed of AI development. Furthermore, we must sprint towards a post-superintelligence world even though we have no real idea what it will bring.
You don't need to do all three. You might not need to do any of them. But if Reddit matters to your audience—and increasingly, AI is making that decision for you—pick one approach and commit to it properly.
Wild Signal provides AI search optimization services that help brands earn citations in the digital landscape. They pair communications experts with geospatial tools to deliver strategic intelligence. via wildsignal.agency
Comprehensive Coverage
139 Skills - Extensive coverage across all major scientific domains
28+ Databases - Direct access to OpenAlex, PubMed, bioRxiv, ChEMBL, UniProt, COSMIC, and more
55+ Python Packages - RDKit, Scanpy, PyTorch Lightning, scikit-learn, BioServices, PennyLane, Qiskit, and others
GitHub repository with 139 scientific skills for Claude, enabling complex multi-step research workflows across biology, chemistry, medicine, and other scientific domains
I started my tools.simonwillison.net site last year as a single location for my growing collection of vibe-coded / AI-assisted HTML+JavaScript tools ... The new browse all by month page shows I built 110 of these in 2025!
Simon Willison's Annual review blog post examining major trends and developments in large language models (LLMs) during 2025.
Subscribe to Newsletter
Stay up to date with email notifications of new posts.
Discussion