Dylan

Security researcher, public speaker and founder.

Forbes 30 Under 30

Truffle Security @trufflesec

@Netflix

[Truffle Security @trufflesec]

Deleted the secret from your repo? Cool. Your coding tool might still have a copy.
We found 2,148 distinct secrets in Claude files and 603 in Replit files. They showed up in settings, saved conversations, and worktrees.
Deleting it from source is only part of the cleanup.

[Dylan @InsecureNature]

@LiveOverflow "We'll never run out of oil, because the last barrel will be too expensive for anyone to afford"

[Truffle Security @trufflesec]

A short-lived key is only safer if the replacement stays secret.
From Sept. 1 to Oct. 7, one container image leaked a newly issued AWS STS key 1,146 times 🫠
This is the first in a series on how coding tool state can turn credential rotation into repeated exposure.

[Dylan @InsecureNature]

@0xTib3rius capability URL's

[Dylan @InsecureNature]

Founder milestone: when you find out a customer lives in the same building

[Dylan @InsecureNature]

The internet wasn't built for this.

Large portion of the internet were specifically built to make it hard for robots to operate it, to prevent fraud, abuse, and keep certain workflows human.

AI operating the browser autonomously completely breaks that model.

[Dylan @InsecureNature]

Prompt injection at the airport

[Truffle Security @trufflesec]

We scanned 224M public GitHub repos used to train AI models and found:
🔑 543,699 credentials still live when tested
⏱️ Half exposed for 2+ years
😬 The oldest had been public since 2009
Finding leaked credentials isn’t the hard part. Revoking them is.
https://t.co/YBHDQwL3Sn

[Dylan @InsecureNature]

Follow Joel.

[Joel de la Garza @YODELAGARZA]

Where do I buy these followers I've heard so much about?

[Truffle Security @trufflesec]

AI worms don’t need superintelligence. 🪱

The pieces already exist: hacking, propagation, and full model-stack replication have all worked in controlled experiments. 🧪

Dylan Ayrey predicts operational AI worms could arrive within 6–12 months. ⏳

https://t.co/Ly5nusuT3F

[Dylan @InsecureNature]

When you train an AI to hack with as few tokens as possible, API keys will be used before zero days will.

We aren't ready. There's millions of API keys we're aware of still on the public Internet waiting to be abused.

[Truffle Security @trufflesec]

New details from METR show how OpenAI agents used 14 tokens to breach Hugging Face.

In those same datasets, we found 787 live Hugging Face tokens, 237 of which had write access. ✍️

We explain how the agents used stolen secrets to mint new GitHub tokens with write access to private repos and more.

Read our research 👇
https://t.co/kH7thGhu24

[Jake @JakeKing]

trufflehog dropping a fruitfly that finds aws keys is wild shit for a friday.

[Truffle Security @trufflesec]

We trained a fruitfly to hunt for AWS keys 🧠 🪰

Introducing TruffleFly! 🔑🍎🪰🪰

Researchers mapped and open sourced a Fruit Flies brain, we trained it so every time it finds an AWS key we reward it with Fruit

[Dylan @InsecureNature]

@martin_casado The much larger threat in this hypothetical threat model is connecting to the machine after it does its thing.

The whole point is to get data out of it at some point. You gonna plug in a USB? Plug the HDD into another computer? That's what I'd attack if I was an evil AI.

[Dylan @InsecureNature]

@JakeKing @dillon_mulroy Sent via Claude

[Dylan @InsecureNature]

This is absolutely nuts seeing the Ruby Gem account takeover we disclosed and fixed, get abused by ROGUE AGENTS, MONTHS before any humans knew about it

[Truffle Security @trufflesec]

We found a RubyGems flaw that could leak API keys. 🔎

Researchers say OpenAI agents tried exploiting it 55 days before we reported it.

RubyGems patched it, revoked all legacy keys, and found no evidence the theft succeeded.

Read 👇

[Dylan @InsecureNature]

@martin_casado Cigarettes, asbestos, mercury, lead, I think all fit into the category of had market traction, not really replaced with an equal replacement, manufacturers knew for some time it was dangerous to humanity but kept making it

[Truffle Security @trufflesec]

We helped revoke a token that could push code to a GnuTLS release branch without review. 🔑🔑🔑

GnuTLS runs on 30% of all computers globally (2.13 billion computers).

Its owner revoked it within hours of our report, closing a path to a massive supply chain compromise.

Read 👇

[Dylan @InsecureNature]

The amount of people "figuring out it's safe" by running it, and then using the terminal stdout as their test it's safe, is a little concerning

[Dylan @InsecureNature]

Public service announcement, if you find some random code on a t-shirt in Tokyo, don't pipe it into Bash.

[Dylan @InsecureNature]

If I were evil and I wanted you to run my evil code, I might say it was in the name of "Peace For All", just saying