In partnership with
News of the day
1. Google's new Gemini 4 Argon model matches GPT-6 Astra but trails Claude Opus 5.5, impacting its market position. → Read more
2. Airbnb CEO Brian Chesky discusses the future of AI interfaces, arguing for a new operating system for AI agents beyond current app structures. → Read more
3. Legato's AI-powered hearing glasses, Legato Frames, are now available for purchase, offering a discreet and accessible solution for moderate hearing loss. → Read more
4. NVIDIA AI factories are engineered to be productive, durable, and fungible, maximizing earning capacity, useful life, and demand for AI hardware investments. → Read more
Our take
Hi Dotikers!
Google finally woke up. After seven months of silence at the frontier, Gemini 4 Argon has landed, and the tech world spent the week arguing about whether it matters. Short answer: yes, but not the way Google wants you to think.
On paper, Argon looks impressive. Google's own comparison table shows it leading or tying on 13 of 18 benchmarks against GPT-6 Astra and Anthropic's Claude models. But here's the catch worth remembering: Google ran nine of those 18 tests itself, and the announcement conveniently names only benchmarks where Argon shines. Independent testing tells a more sober story. Argon matches GPT-6 Astra but still trails Claude Opus 5.5, the current king of the hill. Grading your own homework and then bragging about the results is a bold strategy, even by Silicon Valley standards.
The pricing story deserves scrutiny too. The per-token cost looks attractive until you realize Argon consumes more than twice as many tokens per task as Astra. Cheap tokens that you burn in bulk are not cheap at all, and enterprise buyers who only look at the rate card will learn that lesson the expensive way.
Then there's availability, or the lack of it. Argon is rolling out only to vetted cyber defenders through Google's Fairwind Program, with API access and paid tiers promised for later. Google frames this as responsible, phased deployment aligned with the US government's voluntary pre-release process. Fair enough, but it also means most of us are evaluating a model we cannot touch.
The real takeaway is that the frontier is now a three-way dead heat, and that is genuinely good news. Competition this tight means faster iteration, better prices, and fewer reasons to be locked into one vendor. Argon did not win the race, but it made the race worth watching again.
Alex.
Webinar Series: Running Agents in Customer Work
Every AI vendor reports a resolution rate. No two calculate it the same way, and none of them tell you whether the replies were actually good.
Running Agents in Customer Work is four 30-minute sessions for the leaders who have to make agentic AI work in support, success, and ops. Session 4 gets into how to score quality at volume without reading every transcript, and the questions to ask a vendor about what their resolution math leaves out.
Four Tuesdays, Oct. 13 through Nov. 3, 10 a.m. PT. Register today, one sign-up covers all four.
Meme of the day
