Really good to know. That should have made it into their update letter in point (2). Empowering the user to choose is the right call.

Not marching, but Ukraine uses continuous track machine gun robots seemingly very effectively. They aren’t suicide ones.

https://archive.is/dpNsN

They are an interesting prospect but their use isn't quite as claimed.

They are extremely vulnerable to the same drones humans are.

It's more along the lines of this is a patch were not expecting active fighting this robot can act as a deterrent and surveillance.

Cheaper and simpler than a loitering IRS drone. But more concentrated in domain.

I believe for a while Samsung developed similar drones for the demilitarised zone in Korea. Those could be static as they were hard wired in.

> They are extremely vulnerable to the same drones humans are.

I am not confident about this. Human gets disabled by few small shrapnel projectiles into soft tissue. It is possible to build way more protected robot, for which you need some direct hit to disable it. That robot could also be very agile: e.g. do some evading jump at the last moment before being hit.

I think you just pitched a Robot Wars revival for 2026.

This article shows them being used for offense.

https://edition.cnn.com/2026/04/20/europe/robots-ukraine-bat...

At the moment, cruise speed and manufacturing price.

Solutions like https://bugherd.com/ might make the issue context capture part more accurate.

I spec'd up an implementation of this that uses a hardware button with colors that is in reach of either party. The customer went with a different vendor based on price/"complexity"/training.

We use Jamf Pro for a small company. I'm not a big fan of the minimum 20 seat pricing model. I hope this will be something small companies can move to easily and have enough coverage to satisfy security reviews.

For a less simplistic look at a similar question I'd recommend this article.

https://hyperfocusinhalifax.substack.com/p/why-arent-oil-pri...

The current A/B test I seem to be in is that bad. But it will likely drive the metrics they are trying to drive.

Then just write the extra paragraph rather than bait?

Bait what exactly ? Getting the user to type "yes" ? Great accomplishment.

Sometimes I want the extra paragraph, sometimes I don't. Sometimes I like the suggested follow up, sometimes I don't. Sometimes I have half an hour in front of me to keep digging into a subject, sometimes I don't.

Why should the LLM "just write the extra paragraph" (consuming electricity in the process) to a potential follow up question a user might, or might not, have ? If I write a simple question I hope to get a simple answer, not a whole essay answering stuff I did not explicitly ask for. And If I want to go deeper, typing 3 letters is not exactly a huge cost.

You send all the tokens an extra time at least

I’m not privy to their data on what this does to engagement, but intuitively it seems like the extra inference/token cost this incurs doesn’t align with their current model.

If they were doing it to API customers, sure, but getting the free or flat-rate customers to use more tokens seems counterproductive.

It juices their "engagement" metrics, which is the drug of choice for investors, right up there with net promoter scores.

We’ll see how this plays out. It’s a turbocharged version of enshittification, at a time when other models are showing stronger growth in B2B and other valuable markets.

I canceled my ChatGPT subscription and jumped to Claude, not for silly political theater, but just because the product was better for professional use. Looking at data from Ramp and others, I’m not alone.

Where ORMs are clearly weak is in generating suboptimal queries and making it too easy to create N+1 issues. My first introduction to ORMs was Ruby on Rails. You would rely on New Relic to identify performance issues and then fix them.

With solid AGENTS.md / CLAUDE.md, I do not think this would happen as much anymore in new code. So then it is just a matter of style preference (ORM vs whatever else).

That's exactly why Oxyde has no lazy loading at all. If you don't call .join() or .prefetch(), related data simply won't be there. N+1 is impossible by design, not by discipline.