NYTimes: “METR’s researchers attempted to track this by creating a benchmark of software engineering tasks — like debugging code, setting up servers and training small A.I. models. They hired expert software developers to do the tasks. Then they had A.I. agents attempt the same tasks. When an agent succeeded at a task, they logged the time it had taken the human expert to do the same work. They plotted the results on a single chart — task length on one axis, time on the other — and produced a trend line across years of A.I. progress. What they found was surprising. The length, in human-hours, of a task an A.I. agent was able to complete reliably was doubling roughly every seven months. More recently, with models like Anthropic’s Claude Opus 4.5 and OpenAI’s GPT-5.2, the line took a sharp upward turn — the task length is now doubling every three to four months.”
SaaStr: “The durable moats in B2B in 2026 are not the ones that used to matter. They are: 1. Distribution. Who already has the customer’s trust and attention. 2. Data. Proprietary datasets that compound over time and that competitors genuinely cannot replicate. 3. Network effects. Real ones, not marketing-deck ones. 4. Brand. Actual brand, built over years, that customers trust when they’re nervous. 5. Speed of iteration. The ability to ship 10x faster than the incumbent. (Which, ironically, AI both enables and commoditizes at the same time.) Notice what’s not on that list: your feature set. Your localization footprint. Your integration catalog. Your admin panel. None of that is a moat anymore. None of it.”
WSJ: ““Feed the People!” has two important premises. The first is that nobody should call our food system “broken,” the all-purpose label that armchair revolutionaries slap on any aspect of the messy human condition that doesn’t measure up. “While we agree that there is much wrong with the food system, ‘broken’ is not the correct word to describe it,” the authors write, since the claim “offers no real vision of a better future and only vague gestures at systemic change.” Messrs. Dutkiewicz and Rosenberg sensibly argue that a vast modern society can only be fed safely and affordably by means of an efficient, industrial-scale food-production apparatus—which is exactly what we are fortunate to have. But they want us to work on improving the system to make it more healthful and sustainable and fairer to workers.”
Manu Joseph: “People, irrespective of economic class, have varying capacities for crime. Poverty is just an excuse for those who are predisposed to crime. Most of the poor, like most people, are incapable of serious crime. The entire luxury service industry rests on this fact. If poverty motivates people to steal, whole industries would collapse. The fear of law and perhaps the absence of human rights for the poor do explain a bit of the peace, but not entirely. The only factor that can explain why organized luxury coexists with the poverty of its service staff is that most of the poor do not wish to indulge in mayhem or theft, irrespective of the bad hand life has dealt them.”
