July 2026 Archive
16951.
Stop Prompting, Start Looping
(lanes.sh)
16952.
AI helps scientists design the next generation of medicines
(technologyreview.com)
16953.
LLMs will make scientists do more, less well
(arxiv.org)
16954.
I host my own AIM server
(veronicaexplains.net)
16955.
16956.
Japanese Parody Religion Exists Only to Give Believers a Reason to Say 'No'
(odditycentral.com)
16957.
16958.
Getting Off with the Luddites
(nplusonemag.com)
16959.
The AI Con
(penguin.co.uk)
16960.
16961.
Coolify
(coolify.io)
16962.
Microsoft Is Replacing OpenAI Image Models in PowerPoint, Bing
(bloomberg.com)
16963.
Framework memory costs have doubled
(frame.work)
16964.
Report from the March to Stop the AI Race
(robinsloan.com)
16965.
I regret to report that the AIs are weebs, too
(pcgamer.com)
16966.
The DeepSeek Doctrine
(thexpin.com)
16967.
Emerging challenge: We know students are no longer reading books
(theconversation.com)
16968.
What 4,523 AI investment forecasts revealed about model disagreement
(ipulseai.substack.com)
16969.
ChatGPT Voice is now in the desktop app
(twitter.com)
16970.
China's Moment of Weakness
(foreignaffairs.com)
16972.
How the most-used zero-knowledge proof system (Groth16) works
(blog.zksecurity.xyz)
16973.
16974.
Model Context Protocol prepares to break with its stateful past
(theregister.com)
16975.
The Big Smoke – a living atlas of London
(levels.fyi)
16976.
Show HN: Haserupt – A different take on reducing phone usage
(haserupt.com)
16977.
Doing Okay-ish with AI at the AI-proof competition
(blog.greg.technology)
16978.
Show HN: Country Draw – draw country borders from memory
(countrydraw.net)
16979.
16980.
First experimental evidence of recursive self-improvement (RSI)
(threadreaderapp.com)