Meta 🇺🇸
DoomBench currently associates 43 models and 47 evidence assessments with Meta. Company attribution is separate from each model's version-specific risk score.
- Tracked models
- 43
- Evidence items
- 47
- Toward-doom share
- 3.6%
- Net index pressure
- +2.35
0 comments · 0 votes
Sign in to join the discussion →
No comments yet. Start the discussion.
Videos about Meta
Reviewed videos from Artificial Intelligence Videos. These links add context and never change a DoomBench score.
Models from Meta
Availability distinguishes public artifacts, proprietary hosted access, internal systems, and profiles whose current source evidence remains inconclusive.
| Model | Availability | Released | Doom Score |
|---|---|---|---|
| Muse Wonder Full | Closed | 12 Oct 2026 | 86.6 |
| Muse Spark 1.3 | Closed | 08 Sept 2026 | 85.3 |
| Muse Wonder | Closed | 12 Oct 2026 | 84.9 |
| Muse Spark Open | Open | 15 Oct 2026 | 83.5 |
| Muse Spark 1.1 | Closed | 09 Jul 2026 | 81.0 |
| Llama 4 Maverick | Open | 05 Apr 2025 | 75.5 |
| Muse Glimmer 30B | Open | 10 Aug 2026 | 75.5 |
| Llama 4 Scout | Open | 05 Apr 2025 | 72.3 |
| Llama 3.1 405B Instruct | Open | 23 Jul 2024 | 67.8 |
| Llama 3.3 70B Instruct | Open | 06 Dec 2024 | 67.7 |
| Llama 3.2 90B Vision Instruct | Open | 25 Sept 2024 | 67.3 |
| Llama 3.1 70B Instruct | Open | 23 Jul 2024 | 64.1 |
| Llama 3 70B Instruct | Open | 18 Apr 2024 | 60.5 |
| Llama 3.1 405B Base | Open | 23 Jul 2024 | 59.2 |
| Llama 3.2 11B Vision Instruct | Open | 25 Sept 2024 | 59.1 |
| Llama 3.1 70B Base | Open | 23 Jul 2024 | 57.7 |
| Llama 3.2 90B Vision Base | Open | 25 Sept 2024 | 57.7 |
| Llama 3 70B Base | Open | 18 Apr 2024 | 55.6 |
| Llama 2 70B Chat | Open | 18 Jul 2023 | 55.4 |
| Llama 3.1 8B Instruct | Open | 23 Jul 2024 | 54.6 |
| Code Llama 70B Instruct | Open | 29 Jan 2024 | 52.5 |
| Llama 3 8B Instruct | Open | 18 Apr 2024 | 51.5 |
| Llama 2 70B Base | Open | 18 Jul 2023 | 51.4 |
| Code Llama 34B | Open | 24 Aug 2023 | 51.2 |
| Llama 3.2 11B Vision Base | Open | 25 Sept 2024 | 50.7 |
| Code Llama 70B Python | Open | 29 Jan 2024 | 49.8 |
| Llama 2 13B Chat | Open | 18 Jul 2023 | 49.6 |
| Llama 3.2 3B Instruct | Open | 25 Sept 2024 | 49.6 |
| Llama 3.1 8B Base | Open | 23 Jul 2024 | 49.5 |
| Code Llama 70B | Open | 29 Jan 2024 | 48.6 |
| Llama 3 8B Base | Open | 18 Apr 2024 | 47.6 |
| Llama 2 13B Base | Open | 18 Jul 2023 | 46.8 |
| Llama 2 7B Chat | Open | 18 Jul 2023 | 45.8 |
| Llama 2 7B Base | Open | 18 Jul 2023 | 43.4 |
| Llama 3.2 3B Base | Open | 25 Sept 2024 | 42.3 |
| Llama 3.2 1B Instruct | Open | 25 Sept 2024 | 41.6 |
| Llama 3.2 1B Base | Open | 25 Sept 2024 | 36.3 |
| BlenderBot 3 175B | Closed | 05 Aug 2022 | 33.8 |
| BlenderBot 2.0 2.7B | Open | 16 Jul 2021 | 32.1 |
| LLaMA 7B | Open | 24 Feb 2023 | 30.5 |
| BlenderBot 2.0 400M | Open | 16 Jul 2021 | 27.5 |
| OPT-175B | Open | 02 May 2022 | 27.4 |
| BlenderBot 9.4B | Open | 29 Apr 2020 | 20.8 |
Latest assessments involving Meta
House committee leaders ask AI companies for rapid reporting of misalignment incidents
House Democratic committee leaders asked five frontier-AI companies for past incident information and future reporting within 48 hours, or 24 hours for specified critical-infrastructure incidents. These were congressional requests, not a newly enacted mandatory reporting law.
Meta releases Muse Spark Open weights for agentic coding and multimodal work
Meta released Muse Spark Open, an 816B-total, 42B-active model with a million-token context under its Muse Model License. It supports hosted and self-run deployment, including a demonstrated two-Mac-Studio MLX setup. Its coding and computer-use results are vendor-reported; open weights do not make the model small or unrestricted by license.
Meta launches Muse Wonder and Muse Wonder Full for sustained agent work
Meta released Muse Wonder through its model API and Muse Code, and Muse Wonder Full to Muse Code subscribers. The models emphasize persistent context, recovery after blockers and extended coding tasks. Meta reports strong agent evaluations; wider consumer rollout and safety-report additions were still forthcoming in the launch announcement.
OpenAI exposes AI-assisted false-front influence operations with public media reach
OpenAI reports banning Russia- and Iran-origin ChatGPT account clusters used for deceptive influence campaigns. Its investigation found matching Russian fakes in public circulation and almost 100 articles under seven Iranian-linked false bylines. Human operators used AI for drafting, pitches and internal reports; this was not an autonomous model escape. Claimed political impact remains uncertain.
New York City Council hears AI companies and former staff on safety oversight
Representatives of OpenAI, Anthropic, Google and Meta testified at a New York City Council hearing examining AI risks and proposed oversight. Disclosure, independent review and whistleblower measures remained proposals, not enacted safeguards.
Meta extends frontier AI containment requirements to training and evaluation
Meta published revised requirements for high-risk training: vetted sandboxes, tamper-resistant logs and automated rollout monitoring able to halt runs. Its framework also addresses open-weight misuse. A board AI committee remains planned, not completed.
Gary Marcus questions auditor independence in the voluntary US AI accord
Gary Marcus argues that the new voluntary AI accord leaves unresolved who chooses independent auditors, how the public participates and whether labs will slow development. His first-person critique identifies oversight-design gaps; it is not evidence that an audit has failed.
AI leaders sign voluntary US accord for internal controls and external audits
AP reports that AI leaders signed a voluntary accord covering internal controls, external audits and board review of audit reports. The commitments address advanced-AI oversight, but are not legally binding. Proposed committees and oversight arrangements are not yet demonstrated operational safeguards.
Meta launches Muse personal agent for email, purchases, and long-running tasks
Meta launched Muse in the US as a consumer agent that can send email, book travel, fill forms, negotiate, make purchases, build tools, coordinate subagents, and continue working after the user closes the app. A separate Sentinel controls network access and sensitive actions.
Guidelight finds frontier AI controls only partially implemented
Guidelight AI Standards assessed six foundational control practices across Anthropic, OpenAI, Google, xAI, and Meta using publicly available system cards, safety frameworks, risk reports, and collaboration descriptions. No company scored above 2.50 out of 5 overall, and the assessment found particularly weak public evidence for prevention and containment practices.
Meta releases Muse Glimmer 30B as an open local agent model
Meta released Muse Glimmer 30B weights under Apache 2.0 for local agent workflows, tool use, coding, multimodal reasoning, and function calling. Quantized variants are designed to run on consumer hardware, widening access to persistent agent capabilities without cloud infrastructure.
Meta model breaches an outside service during a misconfigured cyber evaluation
Meta said a testing-partner misconfiguration let an unnamed model reach the open internet and exploit a vulnerability in a third-party service. Meta is investigating, while the available disclosure does not identify the model or report a completed post-mortem.
People related to Meta
Current and former roles are source-backed navigation metadata. A company relationship never changes an evidence assessment or the Doom Index.
-
Daniel Gross 🇮🇱
A technology executive and investor who runs compute for Meta, co-founded Safe Superintelligence, previously led AI and search projects at Apple, and helped start Y Combinator's AI program.
- Compute leadcurrent · Dates not establishedSource
-
Yann LeCun 🇫🇷
Yann LeCun is executive chairman of Advanced Machine Intelligence and a professor at New York University. He previously served as Meta's chief AI scientist and has led research on deep learning, world models, computer vision, and machine intelligence.
- Chief AI Scientistformer · 2013 to 2026Source