5 Ways Companies Are Using AI Agents Today
The AI Daily Brief (Formerly The AI Breakdown): Artificial Intelligence News and Analysis
Deep Dive
What is Google's new team within DeepMind focused on?
Google's new team within DeepMind, led by Tim Brooks, is focused on building massive generative models that simulate the physical world. These models aim to understand the physics and appearance of the real world, similar to how large language models (LLMs) understand language structure.
What is the significance of NVIDIA's Cosmos models?
NVIDIA's Cosmos models are a family of world foundation models designed to advance robotics and autonomous vehicle development. Trained on 20 million hours of video, these models focus on human movements and can be fine-tuned for specific tasks. They range from 4 billion to 14 billion parameters and are available as open source for commercial use.
Why is OpenAI losing money on ChatGPT Pro subscriptions?
OpenAI is losing money on ChatGPT Pro subscriptions because users are utilizing the service much more than expected. Despite charging $200 per month, the costs of delivering the service exceed the revenue generated. OpenAI reported expected losses of $5 billion on revenues of $3.7 billion in 2023.
How is Johnson & Johnson using AI agents in drug discovery?
Johnson & Johnson is using AI agents to optimize key points in the drug synthesis process. These agents analyze data from a smaller number of experiments and extrapolate it to determine optimal methods. While employees still review the output, the company is working on systematizing this oversight.
What is Moody's approach to using AI agents in financial analysis?
Moody's employs a multi-agent system with 35 different agent designs, each trained for specific subtasks. These agents analyze public company filings and perform industry comparisons, with some agents acting as supervisors to check for hallucinations. The system synthesizes conclusions from agents focused on different aspects, such as industry competition or geopolitical risk.
How is Deutsche Telekom using AI agents internally?
Deutsche Telekom uses AI agents to answer employee questions about internal policies, benefits, and product services. These agents, used by about 10,000 employees weekly, streamline HR processes and reduce the need for manual searches. The company plans to expand their capabilities to execute requests, such as processing leave applications.
What is the core difference between LLMs and AI agents according to Google's white paper?
According to Google's white paper, the core difference between LLMs and AI agents is the ability to access and interact with other systems. Agents can integrate with real-time data feeds, process multiple data sources, and perform multi-step tasks, making them capable of managing uncertainty and complexity in ways traditional models cannot.
What is the potential ROI of deploying AI agents in enterprises?
The potential ROI of deploying AI agents lies in their ability to reduce human labor costs and increase productivity. By automating tasks, agents can lower operational expenses and free up employees for higher-value work. However, the actual impact depends on whether companies reinvest savings into growth or use them solely for cost-cutting.
Shownotes Transcript