Microsoft Favors Anthropic Over Openai for Visual Studio Code

This is what I mean - even opinions on domain are wildly different. I've seen people say Claude's React is best.

This is also pretty subjective. I’m a power user of both and tend to prefer Claude’s UI about 70-80% of the time.

I often would use Claude to do a “make it pretty” pass after implementation with GPT-5. I find Claude’s spatial and visual understanding when dealing with frontend to be better.

I am sure others will have the exact opposite experience.

My experience is exactly opposite. Claude excelling in ui, and react. While gpt5 being better on really niche stuff, migth just be me better at caching when gpt5 halucinates as opposed to the claude4 hallucinations.

But after openai started gatekeeping all their new decent models in the api, i will happily refuse to buy more credits, and rather use foss models from other providers (I wish claude had proper no log policies).

Agreed. My experience is GPT5 is significantly better at large-scale planning & architecture (at least for the kind of stuff I care about which is strongly typed functional systems), and then Sonnet is much better at executing the plan. GPT5 is also better at code reviews and finding subtle mistakes if you prompt it well enough, but not totally reliable. Claude Code fills its context window and re-compacts often enough that I have to plan around it, so I'm surprised it's larger than GPT's.

I tried for a while to get Entra working with an external identity provider (Google Workspace).

The other way around worked (Google could use Entra) but it was basically impossible to backend Entra from Google. Weird.

I think people can use stolen / one-time use / prepaid / limited purchase size credit cards fairly easily, too. And you might not find out until after they've racked up a non-trivial amount of costs.

Theoretically yes but a few issues:

- Account creation usually happens before plan selection & payment. Most users start at free, then add a CC later either during on-boarding or after finishing their trial.

- Virtual credit cards are very easy to create. You can signup with credit card with a very low limit and just use the free tier tokens.

Firstly, I can tell you phone number verification made a very meaningful impact. The cost of abuse can be quite high for services with high marginal costs like AI.

Second, all those alternatives you described are also not great for user privacy either. One way or another you have to try to associate requests with an individual entity. Each has its own limitations and downsides, so typically multiple methods are used for different scenarios with the hope that all together its enough of a deterrence.

Having to do abuse prevention is not great for UX and hurts legitimate conversion, I promise you most companies only do it when they reach a point where abuse has become a real problem and sometimes well after.

>Do you have any data on that?

The original stated intention of the practice was that "it" [mandatory phone number registration] "prevents fraud" (though this stance was being critiqued by the person who raised it, not defended).

I'll concede that it probably has stymied some of the most trivial, incompetent fraud attempts made, and possibly reduced a negligible amount of actual fraud, but the idea that it can "prevent" fraud (implying true deterministic blocking, rather than delaying or frustrating) is refutable by the very reasonable assumption that there is almost certainly no company that implements mandatory phone number registration that has or will experience ZERO losses to fraud.

That said, in fairness, this is an unfalsifiable and unverifiable claim, as to my knowledge, there is nothing resembling a public directory of fraud losses experienced by businesses, and there is no incentive for businesses to admit to fraud losses publicly (they may have tax incentives to report it to the IRS, legal incentives to report it to law enforcement, and publicly traded companies may have regulatory incentives to at least indirectly acknowledge operating losses incurred due to fraud in financial reporting), but that doesn't make the claim itself unreasonable or improbable.

>What is to stop a non-fraudulent user from doing the same thing to prevent the tracking by the company?

The argument isn't that mandatory phone registration unavoidably forces privacy infringement upon all users, just that it does infringe upon the privacy of some (I'd suggest a vast majority) of users in practice.

TracFone with minutes at Walmart

If you load it with 10 dollars in minutes, it'll be good forever. But I'm not sure what TracFone has for an inactive policy

I have an Ultra Mobile eSIM as a second line on my iPhone. Costs $3/mo and you can buy cards to top it up or add a credit card. My primary eSIM is a regular phone plan.

It works surprisingly well. I can easily turn off the second line in Settings without removing it.

I could’ve bought an unlocked phone with cash somewhere and used the SIM in that. They wouldn’t know who I was.

I didn’t do that because it was inconvenient and it wouldn’t be anonymous once I started using it for SMS authentication.

Any VOIP provider with SMS support should do the job. I personally use voip.ms but there are many.