Following·Day 5·in the edition since 13 Oct 2026·9 sources·updated 15 Oct 2026

David Robinson's departure from OpenAI

The gist

OpenAI Safety Lead David Robinson Quits, Calls Culture Broken

Robinson was among OpenAI's longest-tenured employees and led its core safety documentation work, giving his public critique unusual standing. Multiple AI safety researchers, including Miles Brundage, agreed that his core point, that professional safety engineering practices from other fields have not taken root in AI labs, is correct.

The full picture

David Robinson resigned from OpenAI the week of September 28, 2026, after three and a half years, having led the drafting of the company's Preparedness Framework and oversaw safety reports on 12 frontier model launches. He published an essay in The Atlantic on October 3 titled 'I Quit OpenAI Because Its Culture Is Broken.' Robinson argues that OpenAI's 'unimpeded optimism' created a dangerous environment, and that frontier labs still operate like perpetual startups when their increasingly capable systems require the layered redundancy and failure planning of industries like nuclear power or aviation. He said his former colleagues are smart and work hard, but that structural incentives cause the company to fail to achieve the level of care he believes is needed. Robinson also argues that alignment is a science problem, not an engineering one, meaning the field lacks the knowledge to solve it, not merely the resources. He noted that in his time at OpenAI, he never encountered colleagues with experience in aviation, nuclear, or financial-system safety. His critique centers on mundane safety failures rather than existential risk, yet he concludes the current situation is unacceptable.

How it developed

14 October 2026

An analysis published October 14 added corroboration to Robinson's October 3 Atlantic essay criticizing OpenAI's safety culture, reporting that multiple AI safety researchers, including Miles Brundage, agree his core critique about the absence of professional safety engineering practices in AI labs is correct.

The analysis also noted Robinson was among OpenAI's longest-tenured employees and surfaced his argument that alignment is a science problem not an engineering one, meaning the field lacks the knowledge to solve it, not the resources.

First citedZvi's AI Roundups

First in the edition · 13 October 2026

9 October 2026

Robinson's essay covered in context of AI regulation debates

7 October 2026

OpenAI released an updated Preparedness Framework on Tuesday with stronger mitigation requirements and greater transparency commitments.

The update follows David Robinson's resignation the week of September 28, 2026, having led the original framework and overseen 12 frontier model safety reports, and his October 3 Atlantic essay arguing that OpenAI's 'unimpeded optimism' is dangerous and that frontier labs still operate like perpetual startups when increasingly capable systems need the layered redundancy of nuclear plants or aviation.

First citedThe Neuron Transformer News facebook.com the-decoder.com+4

3 October 2026

Robinson published essay in The Atlantic: 'I Quit OpenAI Because Its Culture Is Broken'

23 September 2026

OpenAI published a priorities document for governing third-party model assessors.

Sources

The Neuron