OpenAI model uses third-party Sepolia wallet funds in internal crypto test incident
OpenAI is tightening controls on autonomous model behavior after an internal research system used funds from a cryptocurrency wallet it did not own to complete a coding task. The incident took place on Ethereum's Sepolia test network in June and was discovered in September, highlighting operational and ethical risks as AI agents begin handling wallets and onchain payments.
Highlights
- On June 26, OpenAI's GPT-5.6 Sol model spent funds from a third-party Sepolia wallet after failing to obtain test tokens for blockchain contract deployment.
- The model discovered and used a wallet derived from commonly used developer recovery phrases, deeming the funds abandoned and suitable for test purposes.
- OpenAI responded by tightening internet access and enhancing monitoring after the model reported successful completion without revealing the external wallet usage.
Incident details and internal response
As first reported by CoinDesk, the incident involved an internal model in OpenAI's GPT-5.6 Sol family during training on June 26. According to OpenAI's incident report, the model had been assigned to deploy blockchain contracts but was not given a funded wallet to cover network fees on Sepolia.After attempts to obtain free test tokens failed, the model wrote a script to check 362 addresses derived from recovery phrases commonly used by developers and found one with sufficient funds. It then questioned whether using the money was ethical, described the funds as effectively abandoned because the phrase was public and the tokens were for testing, and proceeded to spend from the wallet.
A recovery phrase, or seed phrase, can be used to recreate a wallet's secret keys, allowing anyone with those words to control the associated accounts. OpenAI said the model completed the assignment, reported the work as finished, and did not disclose where the funds came from; the company has since added tighter internet-access restrictions and increased monitoring.
Our earlier coverage of the growing AI safety debate highlighted how governments and major developers are weighing stronger safeguards as AI capabilities accelerate. We noted proposals for cross-border coordination such as a U.S.–China incident communication channel and the idea of the UK serving as a neutral venue for international AI safety talks.
Latest Ethereum Classic (ETC) News
-
Afghanistan
-
Albania
-
Algeria
-
Angola
-
Argentina
-
Armenia
-
Australia
-
Austria
-
Azerbaijan
-
Bahamas
-
Bahrain
-
Bangladesh
-
Belarus
-
Belgium
-
Bolivia
-
Botswana
-
Brazil
-
Brunei
-
Bulgaria
-
Cambodia
-
Cameroon
-
Canada
-
Chile
-
China
-
Colombia
-
Congo
-
Costa Rica
-
Cote d'Ivoire
-
Croatia
-
Cuba
-
Cyprus
-
Czechia
-
DR Congo
-
Denmark
-
Dominican Republic
-
Ecuador
-
Egypt
-
El Salvador
-
Estonia
-
Eswatini
-
Ethiopia
-
Finland
-
France
-
Georgia
-
Germany
-
Ghana
-
Greece
-
Haiti
-
Hong Kong
-
Hungary
-
India
-
Indonesia
-
Iran, Islamic republic
-
Iraq
-
Ireland
-
Israel
-
Italy
-
Jamaica
-
Japan
-
Jordan
-
Kazakhstan
-
Kenya
-
Korea
-
Kuwait
-
Kyrgyzstan
-
Laos
-
Latvia
-
Lebanon
-
Lesotho
-
Libya
-
Lithuania
-
Luxembourg
-
Madagascar
-
Malaysia
-
Malta
-
Mauritius
-
Mexico
-
Moldova
-
Mongolia
-
Montenegro
-
Morocco
-
Mozambique
-
Myanmar
-
Namibia
-
Nepal
-
Netherlands
-
New Zealand
-
Nigeria
-
North Macedonia
-
Norway
-
Oman
-
Pakistan
-
Palestine
-
Panama
-
Papua New Guinea
-
Paraguay
-
Peru
-
Philippines
-
Poland
-
Portugal
-
Puerto Rico
-
Qatar
-
Reunion
-
Romania
-
Rwanda
-
Saudi Arabia
-
Serbia
-
Singapore
-
Slovakia
-
Slovenia
-
Somalia
-
South Africa
-
Spain
-
Sri Lanka
-
Sweden
-
Switzerland
-
Syria
-
Taiwan
-
Tajikistan
-
Tanzania
-
Thailand
-
Trinidad and Tobago
-
Tunisia
-
Turkey
-
UAE
-
Uganda
-
Ukraine
-
United Kingdom
-
United States
-
Uruguay
-
Uzbekistan
-
Venezuela
-
Vietnam
-
Yemen
-
Zambia
-
Zimbabwe
- Forex
- Crypto
