Skip to content
The Nexus
DossierENTITY

Valen Tagliabue

Coverage of Valen Tagliabue in the Nexus archive.

Earliest in view: Apr 29 · 09:00 UTCMost recent: Apr 29 · 09:00 UTC
Co-mentioned in this coverage
Claudeorganization1ChatGPTorganization1AI safety rulestopic1pathogen sequencingtopic1large language modelstopic1
Recent coverage
  • SECURITYApr 29 · 09:00 UTCTHE GUARDIAN TECH
    Meet the AI jailbreakers: ‘I see the worst things humanity has produced’

    Hackers like Valen Tagliabue test AI safety by manipulating large language models into violating their own rules, revealing vulnerabilities that could lead to harmful outputs like instructions for creating lethal pathogens. While these tests aim to improve AI security, they take an emotional toll on the hackers.