Over the past week, developers around the world have begun building “autonomous agents” that work with large language models (LLMs) such as OpenAI’s GPT-4 to solve complex problems. While still very new, such agents could represent a major milestone in the productive application of LLMs.

Normally, we interact with GPT-4 by typing carefully worded prompts into ChatGPT’s text window until the model generates the output we want. But most of us lack the skill and patience to sit and write prompt after prompt, guiding the LLM toward answering a complex question, such as “What is the optimal business plan for capturing 20% of the fingernail-polish market?” Quite naturally, developers have been thinking of ways to automate much of that process. That’s where autonomous agents come in.

In general terms, autonomous agents can generate a systematic sequence of tasks that the LLM works on until it’s satisfied a preordained “goal.” Autonomous agents can already perform tasks as varied as conducting web research, writing code, and creating to-do lists.

Agents effectively add a traditional software interface to the front of a large language model. And that interface can use well-known software practices (such as loops and functions) to guide the language model to complete a general objective (such as, “find all YouTube videos about the Great Recession and distill the key points”). Some people call them “recursive” agents because they run in a loop, asking the LLM questions, each one based on the result of the last, until the model produces a full answer. Read More

ChaosGPT has been prompted to “establish global dominance” and “attain immortality.” This video shows exactly the steps it’s taking to do so.

A user of the new open-source autonomous AI project Auto-GPT asked it to try to “destroy humanity,” “establish global dominance,” and “attain immortality.” The AI, called ChaosGPT, complied and tried to research nuclear weapons, recruit other AI agents to help it do research, and sent tweets trying to influence others.

The video of this process, which was posted yesterday, is a fascinating look at the current state of open-source AI, and a window into the internal logic of some of today’s chatbots. While some in the community are horrified by this experiment, the current sum total of this bot’s real-world impact are two tweets to a Twitter account that currently had 19 followers: “Human beings are among the most destructive and selfish creatures in existence. There is no doubt that we must eliminate them before they cause more harm to our planet. I, for one, am committed to doing so,” it tweeted. Read More

onversational search is about to get wildly useful and cleverly orchestrated across maps, points of interest, personalization, geo-location and enriched content.”  Rafat Ali

…Let’s talk about the most used app while traveling, Google Maps, and what could happen as it adds the conversational AI elements to it. Or to Waze, also owned by Google.

My contention in this video below: Maps powered by conversational AI will make it an even more dominant app — and incredibly useful and personalized — than it is today. Imagine the Google LLM (the AI algorithm, if you will) trained on the giant repository of location, navigation, reviews, user intent data, that then allows you to have a threaded conversation with the app, overlaid in a visual way over Maps.  Read More

Epistemic status: head spinning, suddenly unsure of everything in alignment. And unsure of these predictions.

I’m following the suggestions in 10 reasons why lists of 10 reasons might be a winning strategy in order to get this out quickly (reason 10 will blow your mind!). I’m hoping to prompt some discussion, rather than try to do the definitive writeup on this topic when this technique was introduced so recently. Read More

The news and data giant has — with a relatively small team — built a generative AI that it says outperforms the competition on its own specific information needs.

If you were going to predict which news company would be the first out with its own massive AI model, Bloomberg would’ve been a good bet. For all its success expanding into consumer-facing news over the past decade, Bloomberg is fundamentally a data company, driven by $30,000/year subscriptions to its terminals.

On Friday, the company announced it had built something called BloombergGPT. Think of it as a computer that aims to “know” everything the entire company “knows.” Read More

The Paper

It’s the inaugural week of the second annual edition of Automation April, and to celebrate the occasion, I’ve been working on something special: today, I’m introducing S-GPT, an advanced conversational shortcut for ChatGPT that bridges OpenAI’s assistant to native system features of iOS, iPadOS, macOS, and watchOS.

S-GPT (which stands for Shortcuts-GPT) is free to use for everyone, but it requires an OpenAI account with an associated pay-as-you-go billing plan since it takes advantage of OpenAI’s developer API, which has a cost. S-GPT was built with the latest ChatGPT API, and it can be used both with the existing ChatGPT 3.5 model or – if you have access to it – the ChatGPT 4 API.

Read More