Frontier Daily (Nov 22): 8 items from Hacker News, Latent Space, Simon Willison and more
ClawDOS is an AI coding agent built for DOS, drawing 127 points and 39 comments on Hacker News.
Every day Laojin pulls the official channels and compiles the past 48 hours of AI news into a short brief: curation only, no commentary, with the original link and publish time attached to every item.
1. ClawDOS – An AI coding agent for DOS
ClawDOS is an AI coding agent built for DOS, drawing 127 points and 39 comments on Hacker News.
Source: Hacker News · 2026-11-21 · Original
2. [AINews] Riemann Hypothesis proved...twice independently
Two mathematicians claimed independent proofs of the Riemann Hypothesis within hours of each other: an Anthropic Claude-driven Lean proof and a 940-page fully human proof by Alessandro Masullo.
Source: Latent Space · 2026-11-21 · Original
3. Ask HN: I have been unemployed for 2y. 15 years of experience, US. What to do?
A US senior software engineer with 15 years of experience describes being unemployed for two years and nearly out of savings, prompting 392 points and 415 comments on Hacker News.
Source: Hacker News · 2026-11-20 · Original
4. The War Against Slop and an Emerging Weapon to Fight it: Formal Verification
The article examines formal verification as a defense against AI-generated 'slop', covering companies like Harmonic, Theorem, Axiom Math and newcomer &Truth, whose co-founder Clark Barrett co-created the cvc5 solver used in Amazon's Kiro.
Source: Latent Space · 2026-11-20 · Original
5. Quoting Ravi Vakil
Ravi Vakil says Alessandro Masullo's proposed proof of the Generalized Riemann Hypothesis has been closely reviewed for months by Brian Conrad and looks good so far, with no AI involved.
Source: Simon Willison's Weblog · 2026-11-20 · Original
6. Combatting adversarial distillation
OpenAI describes a layered approach to adversarial distillation that combines model training, safeguards, detection and enforcement to prevent capability extraction.
Source: OpenAI Blog · 2026-11-20 · Original
7. Can AI help scientists identify promising drug candidates?
OpenAI introduces MedChemBench to evaluate how well AI models interpret molecular structures, reason from experimental evidence and select compounds in early drug discovery.
Source: OpenAI Blog · 2026-11-20 · Original
8. PostgreSQL 19 Released
PostgreSQL 19 shipped about two months later than usual amid a wave of AI-reported vulnerabilities, with the author's favorite feature being WAIT FOR LSN for read-your-writes consistency on standbys.
Source: Simon Willison's Weblog · 2026-11-19 · Original
---
_Sources: official blogs of OpenAI, Google DeepMind and Qwen; official RSS of Simon Willison, Latent Space, Import AI and One Useful Thing; Anthropic site monitoring, top AI discussions on Hacker News and open-source updates from key AI orgs on GitHub. Mistral / Meta AI have no stable public feed; X sources are curated by hand. This is a source digest — no commentary from Laojin._
Made by Laojin · AI that ships
365SkillAn agent-skills lab: 14 in-house skills
365Skill is our public lab for agent skills: a standard SKILL.md format, a deny-by-default publish policy, and an evals harness. It holds 14 original 365 skills — 12 public and 2 internal. Apache-2.0 — star it, install it, file issues.
More from Laojin: Sellenca · 365AIOrg · AllModelsAPI · 365Loopa · 365 Ops