Edition 23: Seismic Shifts in AI Safety
Welcome to the AviaryAI Newsletter!
Thanks for joining us as we explore the intersection of GenAI and finance with practical learnings and the latest relevant insights. Let’s get started.
This week you’ll learn:
- ⚖️ California’s AI Legislation Hits the Governor’s Desk
- 🤝 US government teams up with AI giants for safer tech
- 🔬 AI spots diseases earlier than ever at the nanoscale
- 🏦 Learning from Goldman Sach’s first AI deployment
Subscribe to the AviaryAI Newsletter here.
GENERATIVE AI THIS WEEK
The coolest things we're watching and why you should care
California’s Divisive AI Safety Bill Passed to Governor
California lawmakers have passed a groundbreaking AI safety bill, SB 1047, mandating safety testing for advanced AI models and requiring developers to implement kill switches. The bill faces opposition from major tech companies and some politicians, while proponents like Elon Musk support it. Governor Gavin Newsom now has until September 30 to sign or veto the legislation.
So what?
This bill could set off a domino effect of AI regulation nationwide. The new law might lead to a fragmented regulatory landscape, potentially forcing AI companies to navigate a complex patchwork of rules. On the flip side, it could spur the development of more robust, safety-focused AI systems that could become industry standards.
US Gov Partners with OpenAI and Anthropic for AI Safety Research
The two AI giants recently signed agreements with the US government, allowing the newly formed US AI Safety Institute to access and evaluate their AI models before public release. This collaboration aims to ensure safe and trustworthy AI development, with the institute providing feedback on potential safety improvements.
So what?
For financial institutions, the partnership could mean future AI tools come with a government "stamp of approval," simplifying adoption decisions but possibly slowing down innovation. It's a balancing act between safety and progress that will shape the future of AI in finance and beyond.
AI Detects Cancer and Viral Infections at Molecular Level
Researchers have developed AINU, an AI tool that can differentiate cancer cells from normal cells and detect early stages of viral infections by analyzing high-resolution images of cell nuclei. This groundbreaking technology can spot cellular changes as small as 20 nanometers, potentially revolutionizing disease diagnosis and monitoring.
So what?
This breakthrough could transform healthcare by catching diseases much earlier, leading to better outcomes and lower costs. However, there is still important challenges to overcome before this technology is ready for a clinical setting as it requires significant technical expertise and time to set up and maintain the AI systems.
GENERATIVE AI WORD OF THE DAY
Pruning
Pruning in generative AI is like trimming a tree to make it healthier and more efficient. It involves removing unnecessary parts of a large AI model to make it smaller and faster without significantly affecting its performance. This process helps reduce the model's size and computational requirements, making it easier to use on various devices and applications.
GENERATIVE AI IN FINANCE
The latest news at the intersection of GenAI and Finance
Goldman Sachs' First AI Deployment
Goldman Sachs is rolling out its first generative AI tool for code generation across the company, marking a significant step in its AI strategy. The firm has taken a centralized approach, developing an internal platform called the GS AI Platform, which serves as the single point of entry for all generative AI use within the company.
This platform leverages partnerships with tech giants like Microsoft and Google, as well as open-source models, allowing Goldman to fine-tune AI models with its own data while maintaining regulatory compliance and data security.
While the code generation tool is the most widely deployed so far, offering a 20% increase in developer efficiency, the company is also exploring AI applications for document translation and research summarization.
So What?
The centralized platform approach demonstrates a way to balance innovation with security and compliance – critical concerns for financial institutions of all sizes. This model could be adapted for credit unions, allowing them to pool resources and create a shared AI platform tailored to their unique needs and scale.
About AviaryAI
AviaryAI is the next evolution of financial interaction. Enhance your team with proactive outbound voice agents to welcome new members, encourage credit card activations, and drive non-interest revenue.
