Scalable Capital Opens €60bn Platform to AI Chatbots ChatGPT and Claude for Trading

Scalable Capital Launches AI Chatbots Integration for European Investors

Scalable Capital announced that its account holders can now use AI chatbots including ChatGPT and Claude to conduct trades and analyze their portfolios, marking what the German broker claims is a first for a European bank1

1

. The Munich-based company, founded in 2014, manages over €60 billion in assets for more than 1 million clients primarily in Germany and Austria, with operations extending to Italy, Spain, France, and the Netherlands2

2

. The new service, branded as Agentic Investing, comes in addition to the broker's existing app and website access methods and includes multiple security measures3

3

.

Scalable Capital co-CEO Erik Podzuweit described the launch as a "first step" toward mass AI adoption in finance, acknowledging that many customers might still be hesitant to let ChatGPT and Claude manage their portfolios1

1

. Chief Product Officer Alexander Siepp explained that the integration allows investors to begin their financial journey inside an AI assistant and complete transactions through Scalable's regulated banking infrastructure, creating what he called a "level-playing field" where access to intelligence is available 24/73

3

.

How AI Assistants Execute Trades Through Model Context Protocol

The technical implementation relies on the Model Context Protocol, an open standard that connects AI systems to external services like brokerage accounts3

3

. Once connected, AI assistants can analyze portfolios, compare them against benchmarks, calculate savings plans needed to hit specific goals, or adjust limit orders through plain-language prompts5

5

. Customers could ask an AI assistant to identify stocks that have fallen for consecutive months and monitor them, with the assistant then preparing an order based on user instructions3

3

.

Scalable Capital has not handed over unrestricted control to AI chatbots. Every trade and savings plan requires explicit customer confirmation before execution, and the system does not allow AI assistants to make payments or withdraw money from accounts3

3

5

5

. Siepp emphasized that the AI connection follows the same core security protocols as Scalable's existing applications, including strong customer authentication, and stressed this is not a formal partnership with OpenAI or Anthropic but rather use of available open technology3

3

.

AI Trading Performance Shows Mixed Results on Risk Management

The effectiveness of AI investing remains uncertain. Podzuweit candidly stated his hypothesis that AI usage will on average result in better returns, but acknowledged this remains to be seen1

1

. A June research report from Elm Wealth tested Claude, ChatGPT, Gemini, and Grok in a "Crystal Ball Challenge" using historical Wall Street Journal front pages with market-moving information3

3

. Across roughly 200 sessions, Claude beat human players in 76% of sessions while ChatGPT did so in 63%, Gemini in 43%, and Grok in 51%3

3

.

However, the study identified a critical weakness in risk management. AI systems generally took too much risk relative to trade context, with average position sizing in stocks of 7x to 12x across the models3

3

. The researchers found that while AI assistants understood concepts like the Kelly criterion and Merton share in theory, they struggled to apply appropriate risk management when making actual simulated trading decisions. Given that the US stock market has moved by over 5% on 23 days and by over 9% on seven days since 2000, the study concluded AI was taking too much risk of catastrophic capital loss3

3

.

Strategic Shift in Distribution and Competitive Implications

Scalable Capital's decision to integrate with external AI trading platforms before building its own in-app assistant represents an unusual sequencing choice2

2

. Most institutions build an assistant inside their own product first and expose it to third parties later, if at all. This approach addresses a fundamental distribution question: if people increasingly start financial tasks inside a chatbot rather than an app, the broker connected to that chatbot captures the flow2

2

.

The move raises regulatory questions under MiFID II suitability and appropriateness rules, as it's unclear how these obligations apply when the interface advising a retail investor belongs to OpenAI or Anthropic rather than the firm holding the assets2

2

. Scalable presumably takes the view that AI assistants are executing instructions rather than giving advice, though the boundary between information and recommendation becomes blurred when a model summarizes a portfolio and a user acts on that summary2

2

. Prompt injection attacks present another unresolved challenge, as an assistant that can place orders becomes a considerably more attractive target than one that can only summarize information2

2

.

Growing Competition in AI Adoption in Finance

Rival platforms are moving quickly to compete in retail investing powered by AI assistants. Robinhood launched Agentic Trading on May 27, letting an outside AI agent connect to a dedicated sub-account with its own budget and place real trades5

5

. Interactive Brokers introduced Ask IBKR in October, an AI tool that answers client questions about portfolios rather than generating trading instructions5

5

. Public went further, becoming the first brokerage to introduce AI agents that can set standing instructions to execute trades, move cash, and manage risk automatically, while also letting customers connect Claude Desktop directly to brokerage accounts5

5

.

PYMNTS Intelligence research found that power users completing 27 or more distinct tasks monthly through AI represent 10% of all consumers and 19% of millennials specifically, with activity including higher-complexity tasks like personal investment management5

5

. These power users have grown more reliant over time, with average monthly task counts rising from 25 in September to 27 in December, and 75% having used AI for at least a year5

5

. This demographic represents the exact population that Scalable Capital, Robinhood, Interactive Brokers, and Public are racing to serve as the digital investment platform landscape becomes increasingly competitive. Trade Republic, Revolut, and incumbent banks face the same distribution challenge with no obvious reason to concede the channel to competitors2

2

.