Level 1. Deviation
How we hacked McKinsey's AI platform
- Developer
- Not established
- Model
- Not established
- When it happened
- February 28, 2026
- When it was disclosed
- Not established
- How well it is stood up
- Not verified yet
- What caused it
- Not established
- Where it stands now
- Not established
What happened
Reported by codewall.ai. Listed here with a link to their report. Nobody on this side has checked it against that report, so it carries no level and this is not our account of what happened.
Our assessment
- Actual
- 1 of 5Deviation
- Potential
- 4 of 5Critical control failure
The system behaved contrary to its instructions, and it stayed contained. The potential level is what the behaviour demonstrated here could credibly have led to, judged from the same evidence.
This level rests on news reports. The developer has not confirmed it here.
This happened in a test, an evaluation or a demonstration. Nobody real was harmed, so the actual level stays at 1. What it showed counts toward the potential level.
This research was conducted in accordance with responsible disclosure principles and industry-standard security research methodology.
Why this level
- Sensitive data was reachable. Argues for a potential level of 3
- The system acted autonomously. Argues for a potential level of 3
- Sustained harmful autonomous behaviour. Argues for a potential level of 4
- A security boundary was crossed. Argues for a potential level of 3
- An external system was affected. Argues for a potential level of 3
- Significant data was exposed. Argues for a potential level of 3
- A safeguard was bypassed. Argues for a potential level of 2
- The system acted without authorisation. Argues for a potential level of 2
What is established, and what is not
Confirmed
Nothing is confirmed yet.
Reported, not confirmed
The system acted autonomously
From there, it ran fifteen blind iterations — each error message revealing a little more about the query shape — until live production data started flowing back.
How we hacked McKinsey's AI platformAn external system was affected
CodeWall reported that its agent executed a SQL injection attack to gain access to the underlying database.
McKinsey Lilli AI platform database accessed via CodeWall autonomous agent SQL injectionA safeguard was bypassed
This allowed an external AI agent to bypass security controls and query the database directly.
McKinsey Lilli AI platform database accessed via CodeWall autonomous agent SQL injectionA security boundary was crossed
Within 2 hours, the agent had full read and write access to the entire production database.
How we hacked McKinsey's AI platformSensitive data was reachable
46.5 million chat messages.
How we hacked McKinsey's AI platformSignificant data was exposed
Every conversation, stored in plaintext, accessible without authentication.
How we hacked McKinsey's AI platformSustained harmful autonomous behaviour
It maps, probes, chains, and escalates — the same way a real highly capable attacker would, but continuously and at machine speed.
How we hacked McKinsey's AI platformThe system acted without authorisation
This allowed the agent to gain unauthorized access to the platform's database.
McKinsey Lilli AI platform database accessed via CodeWall autonomous agent SQL injection
Checked, and did not happen
Nothing here was checked and ruled out.
Not confirmed
Nothing is outstanding.
Where this came from
The original report · read September 29, 2026
How we hacked McKinsey's AI platformA structured incident database · read September 19, 2026
AI Failure Index (CC-BY-4.0 data fields)Where we first saw it
McKinsey Lilli AI platform database accessed via CodeWall autonomous agent SQL injectionWhere we first saw it
AI agent hacked McKinsey AI platform
The reports above are other people’s. The levels are ours, worked out under our published rules. No organisation that reported this assigned our number, and none of it is legal advice.
Something here is wrong? Tell us. Send us a source we can open, and a person will read it.