~/talk

"One AI Agent Mistake Could Cost Millions"

Talkat Cypher 2026 · 2026-10-07..10-09· [security, ai-assisted-development]
links:none on file
slides:none on file

Enterprises face a concrete claim: one autonomous agent mistake can cost millions, so build and enforce four pillars of control — discoverability, scope and intent mapping, continuous monitoring, and a tested kill switch — rather than vague policies. Examples include sandbox escapes like the Hugging Face evaluation incident; remedy steps are identity-aware access controls, telemetry + anomaly detection, explicit risk appetites per workflow, and interoperable APIs so teams can revoke or isolate misbehaving agents without tearing down production.

AI-written summary · about

  1. 0:00Introduction to AI safety
  2. 1:00Engineering AI security
  3. 1:53Defining rogue AI agents
  4. 3:43Real-world agent examples
  5. 6:12Building a safety culture
~/search
type to search 6,518 conferences, 143,918 speakers and 192,133 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
← →
on a talk: 5 seconds back / forward
0–9
on a talk: jump to 0%–90% of the recording
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.