According to the latest update, we’ll all have AI agents acting on our behalf, doing everything from our weekly grocery shopping to booking airline tickets. It’s gonna change everything, just like the web did! AI is coming for us, they keep saying.
The report highlights that i’ve written before about the trials and tribulations of getting an AI agent to book a dinner reservation. As of now, they’re so slow and clunky, it’s way easier to just fire up OpenTable and do it yourself. So, why do AI agents still seem to suck at everything?
Industry observers note that back then, the nascent “Worldwide scene Wide Web” was very much like today’s so-called “agentic” web – in short, a mess. To understand why AI agents still struggle getting stuff done, just step in a time machine and dial up the year 1995.
The report highlights that welcome to another edition of Prompt Mode, your weekly AI newsletter.
In a fresh development, each week on Prompt Mode, I’ll be serving up analysis of the AI trends that matter to everyday users like you and me. Stay tuned for practical AI tips, hands-on experiences with the most recent AI tools, and–you guessed it–prompts to help you get the most out of your AI assistants. I’m your host, Ben Patterson.
As part of the ongoing story, thanks for reading, and if you like what you see, just sign up right here.
As part of the ongoing story, google was still years away, and staffers at Yahoo were building out web directories by hand. Web crawlers were starting to appear, but there weren’t any agreed-upon standards for web sites to tell those crawlers what they were all about. If you stumbled upon a site, you might not know what it was, who it was for, if it was credible, or whether it was even safe. There were web sites popping up left and right, but no easy way to find them.
According to the latest update, just like users struggled to find URLs back in those days, so do AI agents struggle to find the tools they need to actually get things done online. When you ask an AI agent to book you a dinner reservation and it can’t, that’s a symptom of an agent that can find the right tool for the job. That chaotic, 1995-era state of the web is similar to where we are now with the “agentic” web.
Industry observers note that developed with the backing of Google, Microsoft, Nvidia, and other tech giants, ARD is a standardized “discoverability” layer that allows an online service to say to the global stage, “Here’s what I am, here’s what I do, and here’s my menu of AI tools.” That information could then be crawled, indexed, and ranked by agentic discovery services, which in turn could be searched by AI agents. That brings us to a fresh standard called Agentic Resource Discovery, or ARD for short.
As part of the ongoing story, so yes, we’re essentially talking Google Search for AI, which would make it way easier for an AI agent to — among many other things — find a reputable and agent-ready restaurant reservation service, discover which tools it offers, employ the right authentication methods, and book that table, all on its own.
In a fresh development, google, Microsoft, and Nvidia sure hope so. But if ARD isn’t the answer, something else needs to be. Can ARD help tame today’s chaotic agentic web?
According to the latest update, with their limited context windows, today’s LLMs can only remember so much at once, and once a conversation grows to a certain size, something has to go. Whether you’re using ChatGPT, Claude, or Genini, your favorite AI chatbot will start to forget things if your thread gets too long.
According to the latest update, but there’s also a way to make any AI chatbot “compact” a thread that’s grown too long, allowing you to start fresh in a fresh chat that “reboots” the thread while retaining the essentials of your conversation. All you need is this prompt. Luckily, the best AI tools (like the ChatGPT and Claude clients) can automatically compress lengthy chats.
The report highlights that want more next week? Don’t forget to sign up to start receiving this newsletter in your inbox. Thanks for reading the most recent issue of Prompt Mode.
As part of the ongoing story, his coverage of artificial intelligence interrogates the most recent LLMs, and how they can be used at work and at home to be best prepared for the AI revolution. “AI is going to change our lives sooner than we think,” Ben writes. “Our best way to adapt is by using it every day.” Ben has been a PCWorld author since 2014, and has covered everything from laptops to security cameras before launching PCWorld’s AI beat. Ben's articles have also appeared in PC Magazine, TIME, Wired, CNET, Men's Fitness, Mobile Magazine, and more. Ben holds a master's degree in English literature. Ben has been writing about consumer technology for more than 20 years, and now focuses his reporting on AI as it relates to the basic human experience.