Claude: A Technology Platform for AI Agents in the Industrial Era of Digital Communication and Information Processing (Technical Review of Anthropic’s 2018 Beta Testing Project)
According to Dianne Penn, a project lead at Anthropic, the team has been beta testing the Claude tool since April with a few thousand customers. She anticipates some cool startup solutions: one of the spotlighted customers was Study Fetch, which used it to build a personalized AI tutor called Spark.E.
Other companies are also entering the AI Stone Age. Google demonstrated a handful of prototype AI agents at its I/O developer conference earlier this month, among many other new AI doodads. One agent that dealt with online shopping returns was designed to look for the receipt in the person’s account, fill out the return form, and schedule a package pickup.
Also, this tool can work with images, enabling applications that analyze visual data. Anthropic stated that a virtual interior design consultant can use this tool to process room images and give personalized decor suggestions.
This AI assistant will be available through Anthropic’s Messages API, Amazon Bedrock, and Google Vertex AI. Pricing is based on the volume of text Claude processes, measured in “tokens.” There are usually about 750 words for a 1,000 token. During the beta phase, most users opted for Anthropic’s fastest and most affordable option, Haiku, which costs approximately 25 cents per million input tokens and $1.25 per million output tokens.
The technology has yet to transform white- collar work for the better. Workers are dabbling with chatbots for tasks such as drafting emails, and companies are launching countless experiments, but office work hasn’t undergone a major AI reboot.
Perhaps that’s only because we haven’t given chatbots like Google’s Gemini and OpenAI’s ChatGPT The right tools are limited to taking in and spitting out text via a chat interface. Things might get more interesting in business settings as AI companies start deploying so-called “AI agents,” which can take action by operating other software on a computer or via the internet.
I wrote about the importance of artificial intelligence agents that might prove to be in both the quest to create more intelligent machines and for the drive to make it more useful. Claude’s tool use is a small step toward the goal of developing these more useful AI helpers being launched into the world right now.
Anthropic has been working with several companies to help them build Claude-based helpers for their workers. The online tutoring company Study Fetch has developed a way for Claude to change the look and feel of the platform to fit his needs.
Robot Process Outsourcing: The Challenges of Identifying Interactions and Implications for Return-Bots and other LLMs
Other companies are also cautious, as they haven’t yet launched their return-Bot for use by the mass. Getting artificial intelligence agents to behave is a challenge. The chain of steps needed to complete a task can be broken by incorrect guesses by LLMs who don’t correctly identify what they are being asked to achieve.
Even those early use cases could prove extremely lucrative. The big companies already automate common office tasks with robotic process automation. Recording human workers actions and breaking them down into steps that can be repeated by software is a common method in this area. A lot of work could be automated with agents built on the broad capabilities of LLMs. The analyst firm says the market of robotic process outsourcing is currently worth $29 billion, but expects that to double to $65 billion by year’s end.
