According to CoinMeta, after evaluating production efficiency and risk management issues, Goldman Sachs has prohibited most employees from using advanced artificial intelligence programming tools, a move that has caused dissatisfaction among some staff. In the past few months, Goldman Sachs has allowed its employees to use Claude Code under Anthropic and GitHub from Microsoft in their work. However, according to an internal guideline updated in October, the bank has now restricted the use of these tools to only engineers. Sources familiar with the matter reveal that this ban has displeased some traders, who previously used these tools to speed up tasks such as monitoring client transaction flows, tracking market changes, position modeling, and creating textual and graphical content. This is not the first time Goldman Sachs has tightened restrictions on employees' use of certain AI tools; earlier this year, the bank limited the use of Claude under Anthropic for its employees in Hong Kong. Another source stated that the new directive does not apply to employees known as “strats”, who are mainly engaged in work at the intersection of trading, quantitative research, and software engineering.