Infinite context windows seem to present a very large problem to using AI. Today's models already leak too much old information into current responses, a distraction that is part of why they are cognitively exhausting to use I don't want to work with Borges's Funes the Memorious
RESEARCH
-
Reservoir Computing Enhances Soft Robotics Control
By
–
Unlocking soft #Robotics control with #AI's cousin: Reservoir computing
by Alex Parrish @TechXplore_com Learn more: https://
bit.ly/4f3sTIO #Robots #Engineering #TechForGood #Innovation #Technology -
AI Model Claude Mythos Solves Open Erdős Problem with Novel Approach
By
–
Erdős problem #90 has been open for decades. Over the weekend a mathematician tested whether Claude Mythos could solve it. It did. But what caught my attention: Mythos didn't replicate the known approach from OpenAI's #1196 solution. It repeatedly settled on a different
-
DeepMind CEO Discusses AGI, Curing Diseases, and Post-AGI Human Meaning
By
–
We sat down with DeepMind CEO @demishassabis to find out how close we are to AGI, curing every disease with AI, and human meaning post-AGI.
— The Rundown AI (@TheRundownAI) 26 mai 2026
Full interview below, and the rundown is coming to the newsletter tomorrow! https://t.co/vW3m58yfPPWe sat down with DeepMind CEO @demishassabis to find out how close we are to AGI, curing every disease with AI, and human meaning post-AGI. Full interview below, and the rundown is coming to the newsletter tomorrow!
-
Demis Hassabis on Singularity, AGI Timelines, and Human Meaning
By
–
Google DeepMind CEO Demis Hassabis says we’re in the ‘foothills of the singularity’
— Rowan Cheung (@rowancheung) 26 mai 2026
I sat down with him to talk about what that means, curing every disease, and human meaning post-AGI:
0:00 Intro
0:45 What Demis is most excited about at I/O
1:46 Have AGI timelines shifted?
3:30… pic.twitter.com/Vicpn6GMt9Google DeepMind CEO Demis Hassabis says we’re in the ‘foothills of the singularity’ I sat down with him to talk about what that means, curing every disease, and human meaning post-AGI: 0:00 Intro
0:45 What Demis is most excited about at I/O
1:46 Have AGI timelines shifted?
3:30 -
MiniMax Teases Sparse Attention Architecture for M3 with Significant Speedup Benchmarks
By
–
MiniMax just teased their Sparse Attention architecture for M3. The benchmarks show 9.7x prefilling speedup and 15.6x decoding speedup at 1M tokens vs M2. MiniMax deliberately went back to full attention for M2 because efficient attention wasn't production-ready. Their pretrain
-
AI Agent Token Consumption: Research Paper & Project
By
–
Paper: https://
arxiv.org/abs/2604.22750
Project website: https://
longjubai.github.io/agent_token_co
nsumption/
… -
NPM installations of @openai/codex jump from 100k to 1M per day
By
–
https://npm-stat.com/charts.html?package=openai-codex&from=2026-01-25&to=2026-05-25 … shows that installations of @openai/codex via NPM were about 100,000 per day in January, now over one million per day.
-
Microsoft paper: SkillOpt for self‑evolving agent skills
By
–
One of the most important objects in agentic AI may turn out to be a Markdown file. Not the model weights.
Not the prompt. The skill document. A new Microsoft paper introduces SkillOpt: Executive Strategy for Self-Evolving Agent Skills. The thesis is sharp: If an agent’s -
AI security benchmark separates vulnerability reasoning from file location
By
–
AI security is not one thing. Reasoning about a vulnerability class and locating the responsible file are different problems. Brammert and the team's benchmark separates the two There's gap in every tool tested, including ours Right framing for what comes next in this space