B2BROKER Group has added Model Context Protocol support and an in-terminal AI chat to its B2TRADER platform. The update allows traders to connect external AI tools, including Claude and ChatGPT, to their accounts for analysis and, with the required permissions, trade execution.
London's trading industry is coming home!
Other trading platforms have introduced similar integrations. In September, Leverate introduced an MCP connection that allows traders to use ChatGPT or Claude to analyse accounts and prepare orders, subject to trader confirmation. CMC Markets also connected UK CFD accounts to ChatGPT, but restricted the integration to read-only access.
In August, Finance Magnates reported that Freedom24 had added AI access to its Tradernet platform, while ThinkMarkets and cTrader also adopted MCP-based integrations.
B2BROKER Expands AI From Analysis to Execution
B2BROKER introduced an embedded AI Assistant for B2TRADER earlier this year, offering market forecasts, sentiment indicators, recommendation scores and trading signals. A subsequent update added external market statistics and third-party asset news.
- IG CEO Buys Shares Twice Within a Week After Revenue Outlook Cut; CFO Also Purchases
- FCA Says Only 12% of CFD Firms Changed Charges: How Does Overnight Financing Fees Work?
- NAGA Adds Match2Pay Crypto Deposits and Binance Pay Funding for Non-EEA Clients
The latest release extends these features to account-specific analysis and trade execution . Connected AI agents can retrieve market data, review portfolios and account history, and follow instructions such as closing profitable positions or setting stop-loss orders.
Brokers determine the level of access available to connected agents. Read-only access allows analysis and monitoring but excludes trading functions. Full access permits agents to place, modify and close trades. Brokers can also disable MCP, and each agent is restricted to accounts belonging to the trader who authorised it.
B2BROKER Expects AI Chat to Cut Support Enquiries
The in-terminal AI Chat uses the same MCP layer, allowing traders to ask about markets, positions and platform features. Trading actions proposed through the chat require explicit confirmation. External agents with full access, however, can execute permitted trades without individual order confirmations.
“For brokers, adding AI to a terminal raises an immediate question: who decides what an assistant can actually do?” said Arthur Azizov, Founder and CEO of B2BROKER Group.
B2BROKER expects the in-terminal chat to reduce routine first-line support enquiries. The company also said external AI connectivity could help brokers differentiate their platforms when attracting clients.
Broker AI Integrations Range From Monitoring to Execution
A June report by Finance Magnates Intelligence examined AI integrations across brokers and trading platforms. It found that Claude was supported by nine of the ten integrations studied and compared access models ranging from account monitoring to human-approved orders and AI-driven trade execution.
