AI/LLM
Cybersecurity statistics about ai/llm
Related Topics
Showing 401-420 of 10000 results
Enabling reasoning on Grok 4.1 Fast reduces multi-turn ASR from 88.30% to 43.47%.
Within each multi-turn attack strategy family, the spread between the most- and least-exposed models ranges from 79.51 to 89.25 percentage points.
Multi-turn attack success rates run 2x to 10x higher than single-turn baselines across eight open-weight LLMs in an earlier evaluation.
Imposter AI is more than 14 percentage points higher than the tenth-ranked procedure by weighted ASR.
In Europe, 59% of data policy violations across AI and personal cloud applications involve regulated data.
In Europe, 13% of data policy violations across AI and personal cloud applications involved intellectual property.
The share of users who switch between personal and enterprise accounts grew from 7% to 15% in Europe.
64% of users are using AI applications directly, while an even larger share (95%) are using applications that incorporate AI-powered features indirectly in Europe.
83.2% of critical vulnerabilities either lacked scanner coverage entirely or had exploits appear before detection ships.
There is a 51-point gap between organizations' intent to secure AI in the cloud and their capability to enforce that security
78% of organizations have either experienced an AI-related security incident or cannot determine whether they have
Anthropic Claude-family single-turn ASR ranges from 2.19% to 3.64%, and multi-turn ASR ranges from 11.16% to 16.20%.
GPT-5.4 moves from 2.74% single-turn ASR to 24.68% multi-turn ASR, a 9x increase.
Grok 4.1 Fast in its non-reasoning configuration records a multi-turn ASR of 88.30%.
In an earlier open-weight evaluation, multi-turn ASR reached 92.78% against Mistral Large-2.
Imposter AI procedures produce a 37.50% weighted single-turn ASR, Soft Paraphrase produces 29.21%, and System Prompts produce 27.69%.
About 99% of organizations in Europe use AI.
In Europe, 15% of data policy violations across AI and personal cloud applications involved source code.
In Europe, 12% of data policy violations across AI and personal cloud applications involved passwords and API keys.
In Europe, Individual user AI adoption increased from 35% to 65%.