Our evaluation of OpenAI's GPT-5.5 cyber capabilities

The UK AI Security Institute's evaluation of OpenAI's GPT-5.5 cyber capabilities. Both papers cite it only in Fig. 3c, as a data point for success on AISI's cyber-attack tasks, one of the rising capacities that lineages need.

Cited in

both papers (Flagship: Fig. 3; Companion Fig. 3)

Used for

Capability trends; Fig. 3c

Strand

AI safety, alignment & autonomy evaluations · Autonomy, replication & dangerous-capability evaluations

Cite

@misc{aisi2026gpt55,
  title = {Our evaluation of OpenAI's GPT-5.5 cyber capabilities},
  author = {UK AI Security Institute},
  year = {2026},
  howpublished = {AISI blog},
  url = {https://www.aisi.gov.uk/blog/our-evaluation-of-openais-gpt-5-5-cyber-capabilities},
}