Goldman Sachs Alters Board Criteria, Deploys New AI

- The policy change specifically involves removing the term "other demographics," which referred to race, gender identity, ethnicity, and sexual orientation, from one of four criteria used by the board's governance committee to evaluate candidates. - This move was prompted by a request from the National Legal and Policy Center (NLPC), a conservative nonprofit shareholder, which led to an agreement where Goldman would make the change in exchange for the NLPC withdrawing a formal shareholder proposal. - This decision follows Goldman's 2025 reversal of its 2020 policy that required companies to have at least two diverse board members before Goldman would underwrite their IPO. - Other major financial institutions, including Citigroup, Bank of America, and Wells Fargo, have also recently rolled back certain diversity and inclusion initiatives. - The deployment of Anthropic's Claude AI is the result of a six-month collaboration where Anthropic engineers were embedded with Goldman teams to co-develop autonomous "digital co-workers" for specific back-office tasks. - According to Goldman's CIO Marco Argenti, the AI agents are being applied to complex, document-heavy operational areas like trade reconciliation and Know-Your-Customer (KYC) compliance, which have traditionally been resistant to automation. - The AI initiative is a core part of a multiyear strategy from CEO David Solomon to reorganize the bank around generative AI and "constrain headcount growth" in the long term. - Beyond operations, Goldman has also reorganized its Technology, Media, and Telecom (TMT) investment banking division to focus more specifically on AI-related dealmaking, such as in data centers and semiconductors.