Charles Schwab won't let unauthorized AI agents access client accounts
What's the story
Charles Schwab is blocking unauthorized artificial intelligence (AI) agents from accessing its clients' accounts. The move comes as part of the company's efforts to ensure security and protect client information. Rick Wurster, the CEO of Charles Schwab, said that they are specifically blocking Meta's Muse and other unauthorized AI agents from moving money on behalf of clients.
Security concerns
Risks of sharing passwords with AI agents
In a recent Bloomberg Television interview, Wurster stressed the potential risks of clients sharing their passwords with AI agents like Muse.
He said, "We don't want someone just giving their password to Muse and letting Muse come in and take action on that client's behalf."
The CEO emphasized the need for security measures to protect both clients and the company from potential mishaps caused by unauthorized access.
AI integration
Schwab's proprietary AI assistant
Despite the security concerns, Wurster acknowledged that AI is helping consumers manage their finances.
He said instead of using unauthorized agents with sensitive credentials to move money between accounts, Schwab customers should use Charley.
This is the company's proprietary AI assistant that was launched last month.
The CEO also announced plans for model context protocol servers to securely connect external large-language models with Schwab's systems.
Secure integration
Model context protocol servers for secure access
Wurster explained that these MCP servers would allow Schwab to control the authentication process and what information external AI tools could access.
He said, "We're not going to allow them to move money, but if they want to come in and see their account balances and do research, they will be able to do that through their LLMs."
This way, Schwab is ensuring a balance between security and accessibility for its clients.