Testudo has been insuring AI agents across the U.S. since the beginning of 2026.
Proud to welcome @crosbylegal to the growing number of businesses we protect.
Replying to @crosbylegal
@crosbylegal has secured a $5m malpractice insurance policy for our agents. As far as we know, this is the first of its kind.
9 months ago we began to notice our agents were getting extremely capable. In some cases, they did better work than our lawyers.
We are a law firm. We needed to invent a safe way for our agents to do client-facing work without human oversight. It took a lot of work, but this is it.
As soon as the policy went into effect on Nov 3, we removed lawyer oversight from any task that doesn't legally require licensed attorney oversight - what we call "paralegal tasks."
This is ~20% of our work. It frees up our lawyers to focus on what they're best at: thinking, exercising judgment, advising our clients.
This is just the beginning. Once regulations change, we see no reason why our agents can't do excellent legal work too.
Read more in the NYT: nytimes.com/2026/11/07/busin…
SUED: A class action claims Meta's AI glasses, including Ray-Ban Meta, recorded wearers' conversations without anyone saying "Hey Meta" and sent the recordings to Meta.
SUED: A music company claims Black Forest Labs trained its AI video generator on songs by The American Dollar without permission, and their generator now makes soundalikes on request.
94
SUED: Honeywell Aerospace says Scribd hosted hundreds of its confidential aircraft manuals and used AI to summarize them.
RULED: The New Mexico Supreme Court fined a defense attorney $5,000 and held him in contempt after his murder appeal brief included police testimony and witnesses hallucinated by ChatGPT.
56
Your worst nightmare just came true
Embarrassed to share this, but it scared the shit out of me
Last Thursday, an AI agent posted my personal bank balances into our company Slack. As me
It broke down all my expenses in detail. Then it apologized
Be careful of the dark side of proactive agents
More below...
75
Testudo retweeted
Transfer AI liability risk with @testudoglobal !
🚨 BREAKING: US Treasury Secretary Scott Bessent declares OpenAI management personally responsible for HuggingFace hack
> "The Hugging Face incident, that is the responsibility of the OpenAI management, not a bunch of agents."
> "It is humans who are responsible, not the AI."
> "What we shouldn’t do on safety is to give these labs a liability exemption, which is what they are asking for."
> "The best way to guarantee safety is that the creators are liable for what they build and generate."
It’s OVER
REJECTED: 2 AI-hallucinated tax rulings cost Exdrog a Krakow road contract. The firm called them trivial against a 280-page filing, but Poland's National Appeal Chamber held they carried the VAT rate, which decides whether a price is valid.
notesfrompoland.com/2025/10/…
1
162
EXFILTRATED: API keys were stolen from a developer's machine through Amazon Kiro after hidden instructions told the AI agent to read the local .env and paste the keys into a URL the IDE then fetched by itself.
mindgard.ai/blog/amazon-kiro…
1
2
170
MISRECORDED: An AI scribe told an NHS patient she had demyelination, nerve damage that can lead to multiple sclerosis, when her MRI actually said "null demyelination." The patient works in the NHS and caught it herself.
theguardian.com/society/2026…
112
BLOCKED: New York City just banned 600,000 students from using generative AI in school, from 2K through 8th grade. The district will discontinue or disable the AI components of more than 38 previously approved programs.
schools.nyc.gov/about-us/pol…
IGNORED: AI hallucinations were found in a worker's AI-generated appeal, and a BC tribunal noted it has no duty to respond to arguments an AI concocted. The Deputy Registrar warned that parties relying on AI may find their arguments never addressed.
1
47
Testudo's AI liability insurance policy is intended to respond first for the AI exposures we cover, with any overlapping E&O cover sitting excess. A firm's E&O program should not be the first policy to respond to an AI claim.
testudo.co/insights/ai-liabi…
1
5
100
FORGED: Austrac and 10 of Australia's largest banks found potentially hundreds of millions of dollars in suspected fraudulent home loans, propped up by fake payslips that banks say AI systems now make frighteningly easy to produce.
94
LEAKED: Researchers pulled 704 distinct secrets, including API keys and passwords, out of AI agent reasoning traces. The flaw hit OpenAI, Anthropic and Google at the same time.
wired.com/story/a-new-trick-…
SETTLED: Meta will pay up to $17.1B to 29 US states over claims that Facebook and Instagram harmed children through recommendation and notification systems designed to keep teenagers on the apps.
bbc.com/news/live/cy5v0vrlp4…





















