The vision of "one desktop, many devices" (https://www.divergent-desktop.org/blog/2026/01/26/a12web/#a1... ) seems perfect for cloud hyperscalers to own all of compute. Your desktop will be in the cloud, the only computer with enough CPU power and RAM to run your stuff, and you will be allowed to access your desktop from any device you license from the cloud hyperscaler.
I already have that with my Apple devices, kind of. I can drag my mouse from my MacBook Pro to my iPad, use my iPad (or Vision Pro) as a secondary monitor, or the same with my Mac mini, I can start a document on one platform and continue it on another…
All in all I love the ecosystem, it’s very convenient.
As a separate note, I don't see how A12 Web (https://www.divergent-desktop.org/blog/2026/01/26/a12web/#a1... ) is different from the current web, where (Javascript) apps are downloaded and run locally (in your web browser) all the time. There are some additional checks for digital signatures and package integrity, which are typically taken care of by HTTPS in the current web.
You may think you are not competing. The people whose money you may want (employers, investors, customers) definitely see you as one of many competitors for their funds.
There are two ways to answer your questions. You are asking how do we choose between (1) generate+run (AI generate software for some task, then we run the software to do that task) and (2) agentic execution (AI simply completes the task).
First way to look at this is through the lens of specialization. A software engineer could design and create Emacs, and then a writer could use Emacs to write a top-notch novel. That does not mean that the software engineer could write top-notch novels. Similarly, maybe AI can generate software for any task, but maybe it cannot do that task just as well as the task-specialized software.
Second way to look at this is based on costs. Even if AI is as good as specialized software for a given task, the specialized software will likely be more efficient since it uses direct computation (you know, moving and transforming bits around in the CPU and the memory) instead of GPU or TPU-powered multiplications that emulate the direct computation.
> Similarly, maybe AI can generate software for any task, but maybe it cannot do that task just as well as the task-specialized software.
Yes, maybe, but assuming that is the case in general seems completely arbitrary. Maybe not all jobs are like writing software, but why assume software is especially easy for AI?
> Even if AI is as good as specialized software for a given task, the specialized software will likely be more efficient since it uses direct computation
Right, but surely an AI that can "build pretty much anything" can also figure out that it should write specialised software for itself to make its job faster or cheaper (after all, to "build pretty much anything", it needs to know about optimisation).
Possible, though you eventually run into types of issues that you recall the model just not having before. Like accessing a database or not following the SOP you have it read each time it performs X routine task. There are also patterns that are much less ambiguous like getting caught in loops or failing to execute a script it wrote after ten attempts.
yes but i keep wondering if that's just the game of chance doing its thing
like these models are nondeterministic right? (besides the fact that rng things like top k selection and temperature exist)
say with every prompt there is 2% odds the AI gets it massively wrong. what if i had just lucked out the past couple weeks and now i had a streak of bad luck?
and since my expectations are based on its previous (lucky) performance i now judge it even though it isn't different?
or is it giving you consistenly worse performance, not able to get it right even after clearing context and trying again, on the exact same problem etc?
I’ve had Opus struggle on trivial things that Sonnet 3.5 handled with ease.
It’s not so much that the implementations are bad because the code is bad (the code is bad). It’s that it gets extremely confused and starts to frantically make worse and worse decisions and questioning itself. Editing multiple files, changing its mind and only fixing one or two. Reseting and overriding multiple batches of commits without so much as a second thought and losing days of work (yes, I’ve learned my lesson).
It, the model, can’t even reason with the decisions it’s making from turn to turn. And the more opaque agentic help it’s getting the more I suspect that tasks are being routed to much lesser models (not the ones we’ve chosen via /model or those in our agent definitions) however Anthropic chooses.
> "Is it safe to let an AI into my apps? You approve every connection. You can watch it work in real-time. Nothing sensitive happens without your say. Your data stays yours - we don't train on it."
I worked with (human!) interns and most of them did not require being watched in real-time.
to be fair, the AI-intern may be cheaper than a human intern. And since it is AI, I understand the impulse to require human approval. There's nobody to hold accountable with an AI-intern, so letting it have free reign is scary (to me at least)
reply