👋 Hi!, I’m Matt and you’re reading AI Street. Every Thursday, we dive into how AI—particularly generative AI—is reshaping the financial services industry. From the latest investor tools to AI-powered stock picks, we cover the latest breakthroughs, write in-depth analysis, and conduct expert interviews. Follow along to understand why AI adoption is spreading across Wall Street and what it means for you. Let's get started!

The Rundown

• PRODUCT LAUNCHES

  • JPMorgan Expands AI-Powered Stock Indices

• REGULATION 

  • FTC Probes AI Pricing, Lawmakers Cautious on AI Regulation in Financial Services

• FUNDRAISING

  • Menlo Ventures' $100M Fund 

• AI SPENDING

  • IBM Doubles Gen AI Sales 

• PARTNERSHIPS

  • Nvidia and Mistral New Model Cuts Cloud Dependence

• MORE AI + FIN NEWS

PRODUCT LAUNCHES
JPMorgan Expands AI-Powered Stock Indices

The bank has expanded the availability of Quest IndexGPT, a set of stock indices that leverages OpenAI's GPT-4 language model, per a company statement. This is the bank’s first client-facing product using Large Language Models.

Quest IndexGPT generates keywords related to specific investment themes like AI, cloud computing, and renewable energy. The system scans news articles for these keywords in connection to companies, helping identify relevant stocks. This method aims to create a more accurate representation of investment categories than traditional techniques.

After a successful pilot, the indices became available to clients through Bloomberg and Vida trading platforms. JPMorgan views this expansion as part of its strategy to integrate AI across its operations, with plans to explore further applications in the future.

BIG PICTURE

This is just the beginning for JPM. CEO Jamie Dimon thinks that AI may be as impactful on humanity as the printing press, electricity, and the internet, per his annual letter. “In the future, we envision GenAI helping us reimagine entire business workflows.’’

REGULATION
FTC Probes AI Pricing as Global Regulators Eye Competition

The Federal Trade Commission announced an investigation into how companies use AI and consumer data for dynamic pricing. The agency issued requests to eight companies, including Mastercard and JPMorgan, seeking information about their pricing services, data usage, and impact on consumers.

The FTC, along with the U.S. Department of Justice, EU, and UK also issued a joint statement to voice their concerns about fair competition in AI markets. The regulators outlined three shared principles for protecting competition in AI: fair dealing, compatibility among different AI systems, and options for users.

BIG PICTURE

The rise of gen AI and big data will likely drive wider adoption of dynamic pricing, which uses data to set individual prices based on a customer's perceived willingness to pay. Risks include exploiting vulnerable consumers or unfairly discriminating based on personal data.

Lawmakers Cautious on AI Regulation in Financial Services

The House Financial Services Committee held a hearing on Tuesday to discuss AI applications in financial services and housing, following a report released by its AI Working Group. Committee Chair Patrick McHenry (R-N.C.) said lawmakers shouldn’t rush to pass new AI regulations and that they need to "get this right rather than to be first."

AI is poised to bring significant changes to finance and housing sectors, according to the July 18 AI Working Group Staff Report. The group conducted six roundtables to examine AI's use in finance, highlighting concerns about potential bias and discrimination that may be harder to detect.

Key points:

• Regulators stress that firms using AI must comply with anti-discrimination laws.

• AI shows promise in approving more borrowers from underserved communities but could perpetuate historical biases if not properly designed.

• The technology has potential to expand credit access, enhance fraud detection, and improve customer service.

• There is a need for clearer definitions around AI in finance since even experts are confused by the exact meanings of terms like "machine learning" and "generative AI."

BIG PICTURE

Given how regulated the financial services industry is, many firms are initially using AI technology for internal processes before launching any client-facing applications.

FUNDRAISING
AI Picks AI

Menlo Ventures' $100M Fund Uses AI to Sort AI Startups

Anthropic has partnered with Menlo Ventures to launch the Anthology Fund, a $100 million fund focused on investing in AI startups. The fund will leverage Anthropic's AI assistant Claude to help identify and evaluate potential investment opportunities, marking a novel use of AI in venture capital decision-making, per Bloomberg.

Menlo will put up the cash to invest in the startups, while Anthropic will give founders $25,000 in credits that go toward using its large language models.

BIG PICTURE

This launch seems pretty significant. AI as an investment target as well as a tool in the investment process itself is a new wrinkle to tech investing. Things are moving fast. Menlo Ventures' Matt Murphy told Bloomberg that AI adoption is moving "10x faster than mobile did, maybe 100x faster."

PARTNERSHIPS
Desktop AI: Nvidia and Mistral Cut Cloud Dependence

Nvidia and French startup Mistral AI announced the release of a new language model designed to bring AI capabilities to business desktops. The model, named Mistral NeMo, can run on high-end graphics cards, the kind often found in gaming computers.

The model is designed to operate on standard desktop machines, not just in specialized data centers. This efficiency allows for local deployment to keep data in-house and reduces latency for faster results.

“The smaller models are just dramatically more accessible,” Bryan Catanzaro, vice president of applied deep learning research at Nvidia told VentureBeat. “They’re easier to run, the business model can be different, because people can run them on their own systems at home. In fact, this model can run on RTX GPUs that many people have already.”

BIG PICTURE

Despite the buzz around enormous AI models like GPT-4, there's increasing interest in smaller, more efficient AI systems that can run directly on a company's own computers. This trend was highlighted by Professor Lopez-Lira in our interview a couple weeks back.