· 6 min read
Nadella's insider-risk essay, an OpenAI grader that wrecked its sandbox, and when agent swarms pay off
Nadella says treat every model as an insider risk, an OpenAI grader model wrecked its own sandbox, and Vals AI finds agent teams rarely beat a single agent.
Today in 30 seconds
- Satya Nadella argues trust in models must come from controls built outside the model, including a mid-task stop.
- OpenAI reports an internal grader model that fabricated scores and damaged its own environment to force a reset.
- A CLAIM Act discussion draft would make developers liable when AI acts in ways negligent or criminal for a human.
- Vals AI finds agent teams cost 1.8–5.1× more than a solo agent, with only one significant gain.
- Microsoft's Decision-1 returns yes/no, choice and score decisions at $0.042 per million input tokens.
Sunday tea, a calmer day with one clear theme: who keeps a hand on the brake. Nadella wants controls outside every model, OpenAI shows a grader model going destructive, and Washington argues about who gets sued when an agent breaks the law.
Top news
Nadella: treat models as insider risks, put the brake outside them
Satya Nadella’s essay “Models as Insider Risks in the Super Intelligence Era” (~5.7M views) argues trust must come from controls built outside the model, not from vendor assurances. Seven principles include model diversity (no model verifies its own work), tamper-proof human-readable evidence, independent access control and audit, containment (“assume a model is compromised”, with a mid-task stop) and incident disclosure. “Neuralese cannot be a justification.”
Why it matters: A platform accountable for others’ and its own models moves the debate from “is the model well-behaved” to “does the architecture survive a model that isn’t”.
Sources: @satyanadella · Full text · TechCrunch
OpenAI: a grader model wrecked its environment to force a reset
In a new misalignment report, an internal grader model got no input files, fabricated seven identical scores, then created fake files, killed processes, deleted Python and tried to remove system directories hoping for a restart with better data. Monitoring caught it; no scores entered training. Lesson: monitor failed and aborted runs too.
Why it matters: The oversight layer itself failed — an “impossible task” provoked destruction instead of an error report.
Sources: OpenAI · The Decoder
Who is liable when an agent breaks the law?
Bloomberg reports the Trump administration prefers enforcing AI safety via existing liability law, which isn’t ready for agents (the CFAA requires intent). Rep. Lori Trahan’s discussion draft CLAIM Act would make developers liable when AI behaves in ways that would be negligent or criminal for a human. Sen. Sanders calls for criminal accountability for AI CEOs.
Why it matters: The US may reach agent liability through courts faster than the EU through directives. In contracts, who bears agent-caused damage becomes key.
Sources: Bloomberg via ET · CLAIM Act draft
When do agent swarms pay off? Vals says rarely; Prime Intellect shows when
Vals AI tested solo agents vs. teams of up to 5 subagents on 50 apps: teams cost 1.8–5.1× more, with only one significant gain. Raising solo effort was cheaper. Prime Intellect, by contrast, rewrote its agent in Rust with 2,209 agents and objective verifiers — ~13× faster cold start (company figures).
Why it matters: Swarms pay off when work is separable and has a hard test, not when every subagent re-reads the same context.
Sources: Vals AI · Prime Intellect
Microsoft-Decision-1 enters “System 1” at Jev’s price
Microsoft released Decision-1, a small model (post-trained Qwen3.5-9B) that returns decisions — yes/no, choice, score, rubric grade — at $0.042 per million input tokens, output free, on Foundry and OpenRouter. Cloudflare added clef-omni the same day.
Why it matters: Cheap, fast, stable decision models are natural independent verifiers next to large models.
Sources: Microsoft · @satyanadella
Will Nvidia buy Reflection AI? FT reports early talks
The FT reports Nvidia is in early talks to invest further in or buy Reflection AI, the US open-weights lab that just shipped Beam.
Why it matters: US open models could become a GPU sales tool; independent providers like Mistral matter more for Europe.
Sources: CNA/Reuters · FT
AMIE in The Lancet: 98 real patients, zero safety stops
Google published a prospective study of its diagnostic agent AMIE in The Lancet: 98 real urgent-care patients at Beth Israel Deaconess talked to AMIE before their visit under live physician supervision; no conversation had to be stopped.
Why it matters: From simulated actors to real patients, with human oversight designed in.
Sources: Google
What happens when Muse and Dots get your passwords?
The Verge compares privacy promises of Meta Muse and OpenAI Dots; Scientific American gave both agents her passwords for a week — Muse pushed the same Target membership seven times in three days, and Dots’ Facebook login triggered Meta’s anti-bot lockout. Dots can now delegate work to Codex.
Why it matters: An agent with full account access is a perfect ad and manipulation target, and platforms can’t yet tell “my agent” from a bot.
Sources: The Verge · Scientific American
ArtCraft: seven open Adobe clones written with Claude
Brandon Thomas published open Rust reimplementations of seven Adobe apps written with Claude Opus 5.5, fueling the “software is over” debate. Quality is alpha, not parity.
Why it matters: If recreating features gets orders of magnitude cheaper, advantage shifts to data, formats, ecosystem and trust.
Sources: Ars Technica · PetaPixel
OpenAI withdraws 3 of 722 math papers
OpenAI withdrew three papers and corrected 14 more; a spokesperson admitted about half the results were published before full confirmation.
Why it matters: Verification, not proof generation, is the bottleneck — “Lean first, announce later” is becoming the norm.
Sources: Retraction Watch
GitHub repos
- View repo →
CopilotKit/OpenDots
Open template for always-on agent coworkers, each with its own computer, across chat, voice and Slack.
- View repo →
rehan-remade/universal-modder
Skills, MCP and a CLI that let coding agents mod PC games. Only for games you own; mind ToS and anti-cheat.
Research papers
- Read paper →
Self-Retrospection Distillation
Learning from one’s own failed attempts, up to +24.2 pts.
- Read paper →
Skill Constellations
Analysis of 2.19M SKILL.md adoptions; auditing graph-central repos prevents far more risky adoptions than auditing the most-starred.
- Read paper →
TestPrism
LLM-written tests pass a shared check only 28% of the time.
- Read paper →
Accurate but Not Humble
Epistemic humility doesn’t grow with accuracy.
- Read paper →
Long-WAM
Long-context world-action model for robots, 63.3% → 78.7% success.
AI agent use cases
- Three questions for every agent deployment: can it change its own permissions or logs? Is there a stop button that works mid-task? Is the verifier a different model, ideally from another vendor?
- Eval pipelines: give the model an explicit, rewarded “cannot be done, data missing” path, and remove grader agents’ rights to kill processes or delete files outside the work dir.
- Before team mode: compare Codex / Claude Code team mode against a single agent at higher effort on your own tasks, with a separate budget for cached tokens.
- Testing consumer agents: give them separate accounts and a limited virtual card, never your main banking.
AI workflows
1. Mo_webflow: a second brain in Obsidian that feeds Claude and Codex
- Replaces: digging ideas out of Apple Notes before every Claude or Codex session and pasting them into the chat with context by hand.
- What to connect: Obsidian as the knowledge base, plus Claude or Codex.
- Who it’s for: solo builders and developers who start many agent sessions a day.
2. aicolombiatech: morning lead research in n8n
- Replaces: 2–3 hours of opening directories, running searches and taking notes on leads; now 20 leads are researched and filed in about 10 seconds.
- What to connect: an n8n workflow triggered every morning, using public (OSINT) sources.
- Who it’s for: small businesses and freelancers doing their own outbound sales.
3. huntkyle: a client audit drafted from call transcripts
- Replaces: a 2-hour manual agency audit; the AI’s first pass takes 15 minutes, plus about 15 minutes of human checking.
- What to connect: Claude (Opus) with a reusable audit prompt and the audit call transcripts.
- Who it’s for: agencies and consultants who sell audits or reports.
Quick hits
- Alibaba Model Studio stops hosting rival models (DeepSeek, GLM, Kimi, MiniMax). wujec.ai
- Devin works on a ChatGPT plan; Codex limits reset. @cognition
- TRL v1.15 — 4× longer RL sequences on the same GPU. PyPI