Latest
  1. Johns Hopkins Bayview tells 3,182 patients an employee put their health data into a public AI tool
  2. China’s AI compute access expected to double in 2027, with US-designed chips about 65 per cent
  3. UK commits £100m to national AI Biosecurity Lab
  4. Moonshot AI releases Kimi K3.1 Pro with 1-million-token context window
  5. Residents in Newport and Cardiff oppose hyperscale datacentres near homes
  6. Study: Grok 4.7 picked Republicans in 92% of forced midterm answers, GPT-6.1 Sol picked Democrats in 97%
  7. Attackers exploiting the PHP flaws hand the break-in to AI agents, Unit 42 says
  8. Elvis Presley’s rightsholders sue Anthropic over training on 274 songs

Latest / Frontier models and safety

PolicyFrontierUSConfirmed

OpenAI adopts a beta Preparedness Framework with risk levels and a board override

See how this story connects

Source: OpenAI: Preparedness Framework (Beta)

OpenAI published the beta version of its Preparedness Framework on 18 December 2023. It tracks risks in cybersecurity, CBRN threats, persuasion and model autonomy at four levels, and allows deployment only of models scoring Medium or below after mitigations and further development only of those at High or below. A Safety Advisory Group reviews evidence and recommends actions to leadership, and the board may reverse leadership’s decisions.

Why it matters

A second leading lab tied model release to measured risk thresholds, and its board gained a written power to overturn safety decisions a month after the leadership crisis.

When
18 Dec 2023
ID
WTA-N-2023-0040
Who
OpenAI

Summary by wheresthe.ai, from OpenAI. Checked against the source on 9 Oct 2026.

Read the original at OpenAI

Source copy: download the source PDF as saved on 18 Oct 2026 (PDF). Document the source offers: the source PDF. Also held by the Internet Archive.

Earlier in this Storyline: What led hereAnthropic publishes a Responsible Scaling Policy with AI Safety Levels

Cite this story

WTA-N-2023-0040: “OpenAI adopts a beta Preparedness Framework with risk levels and a board override.” wheresthe.ai, 18 Dec 2023. https://wheresthe.ai/d/openai-preparedness-framework-beta-2023-12/ (source: OpenAI: Preparedness Framework (Beta)).

Spotted an error? Email [email protected].