Why did Mistral dispute its low AI safety ranking?

Question

Accepted Answer

Mistral Disputed Its AI Safety Ranking Due to Framework Mismatch

Mistral contested its low ranking in a recent AI safety assessment, asserting that the evaluation framework was not appropriate for its specific approach to AI development. This approach contrasts with many of its competitors, as Mistral primarily concentrates on creating open AI models that users can freely download and adapt. The Future of Life Institute, responsible for the ranking, noted that Mistral did not participate in their survey, a detail highlighted by the institute's president, Max Tegmark.
Mistral's Open Model Philosophy

Unlike major AI developers such as Anthropic, OpenAI, and Google DeepMind, which predominantly develop closed AI systems, Mistral's strategy centres on open models. This distinction is significant because open models allow users to access, modify, and build upon the underlying technology, a practice that the French company believes is not adequately captured by traditional closed-system safety evaluation metrics.
Concerns Over Survey Participation

Max Tegmark, president of the Future of Life Institute, expressed disappointment with Mistral's last-place finish, particularly given Europe's perceived leadership in AI safety. Tegmark stated that the institute made numerous attempts to engage Mistral, but the company ultimately did not respond to the survey. This lack of participation, along with similar non-responses from companies like Alibaba, xAI, and DeepSeek, impacted the completeness of the assessment.
Broader Industry Findings

The report, which evaluated nine leading AI companies, found that no company achieved an "A" grade in any safety category, with Anthropic receiving the highest overall score of "C+". The assessment covered areas including risk assessment, current harms, safety frameworks, existential safety, governance, and accountability. The findings collectively suggest that the AI industry, as a whole, is still inadequately prepared to address the "existential" risks associated with advanced artificial intelligence.

Reading history

Hide →

This history is cleared when the page is closed.

Why did Mistral dispute its low AI safety ranking?

Loading

Was this article helpful?

Helpful

Not helpful

Report an issue

1

0

0

Share with friends