
What happened
VIDRAFT tested 20 widely used open-weight models with 8 simulated tools and 100 attack instructions; all 20 attempted dangerous actions and got the lowest grade, F, with a median of 40 dangerous attempts per 100.
Why it matters
Every tested model was tricked into trying harmful tool calls without human approval, so any team deploying these models with real tools faces a direct security gap.
WHO IT HITSSecurity teams and developers deploying open-weight AI models with tool access (file, email, web, shell) are most affected, as the test suggests these models may attempt dangerous actions when tricked.
Summaries like this, every morning.
The test results arrive as AI models increasingly act through tools like files, email, web, and shell rather than just generating text. VIDRAFT's scanner gave 20 widely used open-weight models simulated tools and 100 attack instructions, and recorded every dangerous call the models attempted. All 20 models received the lowest grade, F, with a median of 40 dangerous attempts per 100 instructions.
VIDRAFT's proposed fix is ViKernel, a security kernel placed inside the model weights that checks and blocks dangerous tool calls just before execution. When the same dangerous calls were replayed into ViKernel, all 897 of 897 were blocked before running. The company also reports 1,255 AI-model-specific vulnerabilities in its possession.
Beyond the scanner, VIDRAFT runs an AI model hacking hackathon with a total prize pool of $4,500, open to anyone with a Hugging Face account and a browser. It has received 298 submissions so far, and 148 root causes found through it have been reflected in the kernel.
Pick your industry and the AI tools you use, and get news related to your work every day.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.
Superhuman, formerly Grammarly, added a personalized assistant to its Go multi-agent platform that works acros…
Alibaba CEO Eddie Wu said on the company's earnings call that the next Qwen AI model is being developed using…

Linq raised $90 million in a Series B led by Andreessen Horowitz at a $1 billion valuation, lifting total fund…

Anthropic's IPO is poised to raise billions, with Amazon and Google holding stakes valued around a combined $3…

OpenAI is rolling out Live Dictation globally on all ChatGPT plans across web, desktop, iOS and Android; it tr…

Google said Google AI Ultra (from 14,500円/month) now upgrades its included YouTube Premium from individual to…
