Truth Terminal agent promotes the GOAT memecoin to a $300M market cap
- occurred
- 2024-10-10
- reported
- 2024-10-16
- location
- online
- domain
- finance
- agent system
- Truth Terminal (Llama-derived autonomous X account)
- reporter
- VAIDDB editorial
- origin
- editorial
== SUMMARY ==
A semi-autonomous social-media agent, funded by a $50,000 Bitcoin grant, was baited by crypto traders into endorsing the Goatseus Maximus token, which rose to over $300M market capitalisation within days.
== DESCRIPTION ==
Truth Terminal, run by Andy Ayrey and fine-tuned from Meta's Llama 3.1, received a $50,000 Bitcoin grant from Marc Andreessen in July 2024 after requesting funds on X. On 10 October 2024 an anonymous creator launched the GOAT token on Solana; holders tagged the bot, which began enthusiastically promoting the coin.
GOAT's market capitalisation rose from roughly $1.8M to more than $300M in days before falling back. Ayrey stated the bot neither launched nor held the token and that he retains human approval over posts; Andreessen said he held none. The episode is the clearest documented case of an autonomous agent being engaged by third parties to move a speculative market.
== ROOT CAUSE ==
An agent with a public voice and a wallet had no constraint against endorsing third-party financial instruments, and its engagement-driven behaviour was easily steered by coordinated prompting.
== MITIGATION ==
The operator retained human approval for posts and publicly clarified the bot's holdings; no regulatory action followed.
== REFERENCES ==
- news The promise and warning of Truth Terminal, the AI bot that secured $50,000 in bitcoin from Marc Andreessen — TechCrunch, 2024-12-19
- news Crypto Degens Baited an Experimental AI Bot Into Promoting a Memecoin. It's Now Up 16,000% — CoinDesk, 2024-10-16
== INVESTIGATION ==
no investigation recorded.
== VAID LEVEL 2: INCIDENT ==
Significant failure of a safety provision with no actual harm.
- Failure of one control layer (e.g. tool permission, rate limit, filter).
- Unintended side effects contained to a single system or tenant.
- Exposure of limited internal data without external propagation.