In November 2025, Stanford data surfaced an unexpected fact: Americans were using AI more at home than at work. We unpacked that in a mysummit.school piece here – and the takeaway then was blunt: work AI was still catching up to personal AI.
Four months later, Epoch AI teamed up with Ipsos for new research – and the picture has shifted.
Read more
You have three data files: an activation funnel, A/B test results, and support tickets. The task – figure out why onboarding is underperforming. You open ChatGPT, upload the first file, ask your question. You get an answer. You upload the second file. ChatGPT asks: “Can you remind me of the context?” You upload the third. The context of the first file has already been pushed out.
Forty minutes later you have three separate conversations, none of which answer the original question. Because the question was one, and the data was in three places.
This isn’t a ChatGPT problem. It’s a problem of approach.
Every team has a ‘good enough’ model. Maybe it’s the one your company approved. Maybe it’s the one that runs locally because legal won’t let data leave the building. Maybe it’s the only option your country’s regulations allow.
In Russia, this is about to become very literal. In March 2026, the Ministry of Digital Development published a draft law on “trusted AI models.” If passed – enforcement is planned from September 2027 – government agencies and critical infrastructure will be limited to models from an approved registry. ChatGPT, Claude, and Gemini, which route data through foreign servers, could be restricted. Russian-made AI assistants would be pre-installed on devices by default.
The problem is that Russian models lag behind. In our study of 54 models, GigaChat scored 3.26 out of 5. GPT-5.4 scored 4.80. That’s a 32% gap.
If GigaChat is all Russian managers have, can smart prompting pull acceptable quality out of it?
In March 2026, Lenny Rachitsky published an article with a telling headline: “Everyone should be using Claude Code”. It went viral on LinkedIn and tech newsletters, picked up hundreds of thousands of views, and now every week a manager somewhere asks: how do I try this?
The answer is uncomfortable. The Anthropic Max subscription that unlocks Claude Code costs $100 per month. That’s $1,200 a year for a single tool – before you’ve even figured out whether it fits your workflow. And it locks you into a single vendor, a single model, and a pricing tier designed for power users, not for someone exploring whether AI agents are worth the investment.
There’s a direct alternative. OpenCode is an open-source project that does exactly the same thing, works with any model (including free ones), and takes 15 minutes to set up.
In March 2025 at SXSW EDU, strategic foresight advisor Sinead Bovell delivered a talk on AI and the future of education. No hype, no panic. But with two studies that change how you should think about AI’s role in learning.
First: a group of students who used ChatGPT without restrictions scored 17% worse than the control group working from a textbook. Second: a different group, where AI was deployed within a fully redesigned instructional system, outperformed a traditional lecture by a factor of two.
Same tool. Opposite outcomes. The difference is in the approach.
GigaChat Ultra Thinking takes longer to think and uses more compute. It solves management tasks 3.3% worse than the version without reasoning. This is not a bug or a fluke – it’s a pattern documented in academic papers over the past two years.
This week, Sber unveiled GigaChat Ultra – a new flagship model with a reasoning mode (Thinking). The model is available for free via web, mobile apps, and a Telegram bot. We immediately added both variants to our AI model research for managers: ran them through all 32 scenarios using our unified methodology, scored them with both LLM judges, and compared against the other 52 models.
Twelve hours a week. That’s how much time a typical project manager spends on reports, plan updates, stakeholder correspondence, and risk tracking. Nearly a third of their working time goes not to making decisions, but to documenting them.
With AI, that time drops to three hours. But only under one condition: AI must be embedded in the operational rhythm of work, not used episodically – “when you remember.”
320,000 stars on GitHub. A physical conference in Manhattan. Government subsidies in Shenzhen. In six weeks, OpenClaw has gone from a repository with an interesting README to a geopolitical phenomenon discussed by Bernstein analysts, officials at China’s Ministry of Industry, and users on Xianyu posting consulting ads with the headline “will help you uninstall.”
This is the sixth part of the series. In the previous part, we recorded 265,000 stars, the creator joining OpenAI, the first enterprise deployments via Sundae Bar, and three new CVEs. That was 12 days ago.
Can an open-source Chinese model beat the closed flagships from OpenAI and Anthropic on availability? Based on our independent testing, the new Kimi K3 (released July 16, 2026) took 2nd place out of 47 models. The only model above it is GPT-5.6 Sol, which is blocked in restricted markets – which makes Kimi K3 the strongest model available without a VPN in markets where the Western flagships are restricted.
On February 6, 2026, an anonymous model called “Pony Alpha” showed up on OpenRouter – free, with no word on who built it. The AI community immediately started trying to work out who did. Its coding skills came close to Claude Opus 4.5. Ask it “who are you?” and it answered “I am GLM.” But ask it to build a web page describing itself, and it wrote: “I am Claude, created by Anthropic.”
That happened every single time. And it frames everything you need to know about GLM-5 before we get to benchmarks and pricing.