Red-teaming
Deliberately attacking a model to find its failures before the public does.
Borrowed from military and cybersecurity practice, red-teaming means assigning people (or other models) to coax a system into misbehaving: revealing private data, producing disallowed content, or following malicious instructions. The failures found this way feed back into training and safeguards. Major labs now report red-teaming results as part of a model’s release, and regulators have begun to ask for them.
Where it mattered in the record
-
Reflection AI launches Beam
On 5 October 2026 Reflection AI introduced Beam, its 501B-parameter open-weight model with 23B active parameters and 1M tokens of effective context. Beam stays in final red-teaming; weights will follow this month.