[dev_l1x_be]

What a series of disasters that are happening at Anthropic nowadays. I am not even sure what is going on with Opus 4.7 I had to switch back to 4.6 and 4.6 was already a downgrade (anecdotal + the github thread with the harness changes).

I am cancelling my subscription as it is impossible to justify these degradations and paying for a subpar service especially now that we have at least 3 more models that are as good as Opus and there is the pi project that is undoubtedly the best harness.

[orphea]

I guess this is what you get when you replace common sense with LLMs.

[callamdelaney]

Too much vibe coding

[jeffrallen]

You're absolutely right!

[kilroy123]

Makes me very glad there’s so much tough competition. OpenAI is a flawed company but I’m glad they’re keeping up.

[yayadarsh]

absolute masterclass in shooting yourself in the foot over the past month or two.

[pluc]

I don't know why anyone is still using this - much less defending them..

In the last MONTH, I've asked how you can defend implementing (or even choose implementing) AI when:

the AI you have implemented throughout your company changes the results you've come to trust? https://www.theregister.com/2026/04/13/claude_outage_quality...

or won't let you log in?: https://github.com/anthropics/claude-code/issues/44257

or makes stuff up?: https://dwyer.co.za/static/claude-mixes-up-who-said-what-and...

or when it's down?: https://status.claude.com/incidents/6jd2m42f8mld

or when you get banned?: https://bannedbyanthropic.com/

or installs spyware: https://www.thatprivacyguy.com/blog/anthropic-spyware/

or takes the features you use out of the plan you subscribe to without notice? https://www.theregister.com/2026/04/22/anthropic_removes_cla...

or renders your IP legally unenforceable? https://legallayer.substack.com/p/who-owns-the-claude-code-w...

or stealthily changes pricing terms based on... file names you have? https://github.com/anthropics/claude-code/issues/53262

or invoices you for usage you did not perform, and won't answer support requests until you raise hell on social media? https://nickvecchioni.github.io/thoughts/2026/04/08/anthropi...

i mean seriously, why on earth would you use this? i thought we were professionals

[michaelmior]

The problem in most of those cases is not specifically AI. Many of the issues you cited are related to Anthropic specifically and many could have been avoided with better testing.

[pluc]

Yes, I am assuming the AI/LLM of choice you've implemented in your software engineering org is Claude because as far as I can tell there aren't really alternatives that come close to its quality in software.

[michaelmior]

We have some very heavy users of Codex in my org and we're very happy with the quality (politics aside).

[pluc]

So your thinking in light of this information is just "don't use Claude"?

[michaelmior]

I don't think it's that simple, but I do think a lot of the problems mentioned are not inherent to the use of AI.

[yard2010]

Excuse me for being blunt but you would assume ai bros run a place like this, and ai bros can manage tech as much as crypto bros can manage monetary systems.

On the other hand they make good products.