PulseAugur
EN
LIVE 14:23:10

Researcher poisons open-weight AI model for under $100

A cybersecurity researcher has demonstrated how to poison an open-weight AI model for less than $100. Katie Paxton-Fear, a lecturer at Manchester Metropolitan University, was able to install a backdoor in a model within an hour. This highlights a potential security vulnerability in open-weight AI models, as they may lack robust verification mechanisms. AI

IMPACT Highlights potential security risks in open-weight AI models, necessitating improved verification methods.

RANK_REASON Research demonstration of a security vulnerability in open-weight AI models.

Read on The Register — AI →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

Researcher poisons open-weight AI model for under $100

COVERAGE [3]

  1. The Register — AI TIER_1 English(EN) ·

    Researcher poisons open-weight AI model for under $100

    Models demand trust without offering verification

  2. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    Reasons you still need a human in the loop: "Researcher poisons open-weight AI model for under $100 https:// share.google/6GPUoUdIMJG8EFPWh " Besides the occasi

    Reasons you still need a human in the loop: "Researcher poisons open-weight AI model for under $100 https:// share.google/6GPUoUdIMJG8EFPWh " Besides the occasional hallucinations, the lack of creativity, and an inability to encode certain types of info into the models (at least …

  3. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    The Register: Researcher poisons open-weight AI model for under $100. “Katie Paxton-Fear, a lecturer in cybersecurity at Manchester Metropolitan University and

    The Register: Researcher poisons open-weight AI model for under $100. “Katie Paxton-Fear, a lecturer in cybersecurity at Manchester Metropolitan University and staff security advocate at Semgrep, managed to install a backdoor in an open-weight AI model in about an hour for less t…