A week in Generative AI: Atlas, Memory & Mico

News for the week ending 26th October 2025

Some interesting news this week from OpenAI, Anthropic, and Microsoft. OpenAI released their much anticipated ChatGPT Atlas web browser, Anthropic added memory to Claude, and Microsoft released Mico, a Clippy for the generative AI era…. Ahem…

Lots of Ethics news, as usual with reports from SAG-AFTRA that OpenAI is taking their Sora deepfake concerns seriously, Channel 4 Tilly Norwooding itself, as well as Netflix apparently going ‘all in’ on generative AI.

There’s a great Long Read on Artificial Ignorance, The Field Guide to AI Slop and Ethan Mollick has updated his Opinionated Guide to Using AI Right Now.

---

OpenAI releases their AI-powered browser ChatGPT Atlas

This week, OpenAI released its AI-powered browser, ChatGPT Atlas, to anyone with a Mac. I think AI-powered browsers will be a big deal, and they’re the best example we currently have of ‘zero-click’ browsing, one of the three ‘zero-click’ pillars of Web 4.0.

Web browsers are where the majority of people spend their time on desktop computers and having an always present AI companion will be very helpful, even if its just there to gather context to help with future chats.

However, my experience so far with ChatGPT Atlas, Perplexity’s Comet, and Claude in Chrome is that they’re currently a solution in search of a problem. I just don’t have a good use case for them yet, and all the use cases in the marketing materials are a little bit hackneyed.

If you’re interested in reading more about the launch there are a couple of good articles below:

• OpenAI’s Atlas is more about ChatGPT than the web

• Two days after OpenAI’s Atlas, Microsoft relaunches a nearly identical AI browser

• OpenAI’s New Browser Raises ‘Insurmountably High’ Security Concerns

Source

---

AI writing hasn’t overwhelmed the web yet

Hat tip to

Charlie Guo

for featuring this report, which caught my eye, in his great Long Read on The Field Guide to AI Slop.

Whilst AI generated content has now (slightly) surpassed human generated content online, its rapid growth has now plateaued, making it look like we’re in a new 50/50 status quo. Hopefully it stays there for the foreseeable!

Some other interesting things mentioned in the report is AI generated content has become much better, meaning people often struggle to tell the difference. I’m not 100% sure I buy this - I recognise that I’m in a small minority of heavy users of AI platforms, but I can definitely spot tell-tale-signs of AI generated content, and I think as we see more adoption many more people will be able to spot it too. I think you just start to get a feel for it over time.

Source

---

Claude now has memory

Anthropic has taken their time with adding memory to Claude, but for good reason. Current implementations on other platforms are a bit one dimensional and overly simplistic.

This is why I’m a big fan of how Anthropic are approaching memory in Claude - specific memories can be toggled on and off and edited with natural language. The other big feature is being able to create “distinct memory spaces” that would allow users to keep personal and professional memories separate, for example. Users can also manually import memories from other platforms and also export their memories at any time.

Hopefully all these features will become a bit more streamlined over time, but its great to see some innovation in this space, which will hopefully see similar improvements on other platforms.

Source

---

Microsoft’s Mico is a ‘Clippy’ for the AI era

Oh no…. please no… 🤓

Microsoft Clippy was a bad user experience - it was intrusive, irrelevant and patronising. I’m a big believer that our current AI chatbots need to become more proactive to realise their full potential, so I just hope that Microsoft gets the balance right with Mico this time around.

Source

---

Web 4.0

• Meta AI’s app downloads and daily users spiked after launch of ‘Vibes’ AI video feed

• Cloudflare CEO Matthew Prince is pushing UK regulator to unbundle Google’s search and AI crawlers

• How Ramp Increased AI Brand Visibility 7x in Accounts Payable

---

AI Ethics News

• Bryan Cranston and SAG-AFTRA say OpenAI is taking their deepfake concerns seriously

• Joke’s on you, fleshbag! Channel 4’s first AI presenter is dizzyingly grim on so many levels

• Netflix goes ‘all in’ on generative AI as entertainment industry remains divided

• Anthropic CEO claps back after Trump officials accuse firm of AI fear-mongering

• Several users reportedly complain to FTC that ChatGPT is causing psychological harm

• ‘Sycophantic’ AI chatbots tell users what they want to hear, study shows

• Reddit sues Perplexity for allegedly ripping its content to feed AI

• Amazon claims the headline isn’t robots taking jobs as it reveals new cost-cutting robots

---

Long Reads

• Artificial Ignorance - The Field Guide to AI Slop

• One Useful Thing - An Opinionated Guide to Using AI Right Now

---

“The future is already here, it’s just not evenly distributed.“

William Gibson