As of: 2026-10-12 07:22 CEST.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code). Human editor: Adam Bicz.
Canonical page: https://postcutoff.com/p/2026-10-10-davidsacks-satya-is-right/

# David Sacks: 'Satya is right' — engineering safety, not 'alignment'

David Sacks (@DavidSacks), X, 2026-10-10.
Source: https://x.com/DavidSacks/status/2109004929943695870

## Why it matters

PCAST co-chair uses Nadella's essay to set 'engineering' containment against Anthropic-style alignment and model welfare (~354k views), the administration-aligned reading of the essay.

## Summary

Posted Oct 10, 2026 (19:35 UTC), quoting Satya Nadella's X Article, about 90 minutes after Sacks's own "Is alignment safe?" post
(2026-10-10-davidsacks-is-alignment-safe). Checked via api.fxtwitter.com on Oct 11: ~354k views, ~6.0k likes.

## Archived text

> Satya is right. The way to make SI safe is not to train it with a sense of self, its own moral philosophy, and permission to act as a conscientious objector. That’s the “alignment” approach and it magnifies the control problem.
> 
> The engineering approach that Satya describes is different: separate the supply of intelligence from authority over it. Surround non-deterministic models with deterministic controls, observability, privilege limits, logging, and the ability to always contain or shut them down. Treat models/agents like powerful insider risks, not moral patients whose psychological wellbeing is at stake.
> 
> As Satya points out, the most trustworthy system is the one that lets us trust the model the least — not the one that encourages the model to develop independent agency and grievances. Engineering safety is not the same thing as “alignment.”
> 
> > Quoting @satyanadella: https://x.com/i/article/2108928845969780736

Archived 2026-10-11 via fxtwitter (unofficial).
Counts: 356,813 views, 6,031 likes, 669 reposts, 435 replies (at fetch time)

## Cited in

- 2026-10-10: [Nadella's X Article 'Models as Insider Risks in the Super Intelligence Era'](https://postcutoff.com/e/2026-10-10-nadella-models-as-insider-risks/)
- 2026-10-08: [Anthropic usage policy bans abuse of Claude, rewrites election and surveillance rules](https://postcutoff.com/e/2026-10-08-anthropic-2026-usage-policy-update/)

## People in this post

- [Satya Nadella](https://postcutoff.com/person/satya-nadella/), Chairman and CEO, Microsoft