Anthropic bans orgs without warning | Hacker News Reader

Anthropic bans orgs without warning

48 pointsalpinisme20 comments

I work at at an agricultural technology company. On Monday, everyone in our org woke up to emails saying that their Claude accounts had been suspended (~70 users).

At first -- since the email was to me, with a link to a Google Form if I personally wanted to appeal -- I thought it must be an individualized ban (at least after deciding it wasn’t a phishing attempt). I couldn’t figure out why, but it set me searching my mind for possible triggers in my recent activity.

On Slack, though, it quickly became apparent this was actually an organization-wide ban. And none of us had been warned, including our account admins. We submitted the Google Form, but that was just a black hole. We’re waiting to hear back still a day and a half later.

But this is insane for a number of reasons:

1. Banning an organization for the behavior of an individual is a recipe for disaster in a business context. Disgruntled employees, incompetent interns -- anyone could maliciously or accidentally revoke Claude access for the whole business.

2. We didn’t just have a Claude Team plan, we also had an API account, which is paid for separately but had the same admins. The API account continues to allow us to use our API keys and sent us a renewal bill yesterday (after the Team account suspension). But none of our admins can actually view usage or billing, because our email addresses were banned.

3. Banning without warning makes every move dangerous. Was it because we had conversations about fertilizer? GPS satellites? other agriculture-related things? We can’t know and can’t avoid it.

We’ve reached out to Anthropic via a number of channels but have received only radio silence. There was a twitter thread about a similar issue (https://x.com/patomolina/status/2045281665363386504), and we tried DM’ing the Anthropic employee who chimed in there. Also no response. I’m sure if we wait long enough we’ll come to some form of resolution here, but you have to ask yourself if this is a platform you can entrust your daily workflows to as a business.

> you have to ask yourself if this is a platform you can entrust your daily workflows to as a business.

No, it isn't. No LLM platform ever will be. No platform or vendor of any kind ever will be, if we are being honest. One cannot set up a business where another company becomes critical to your operations. You can certainly use platforms and vendors in your day-to-day operations, but you always need a backup / business continuity plan because you never know when a vendor will flake out on you, for any variety of reasons.

It seems like many startups learn this lesson painfully, and most people who have been around the block a few times know it well. So I'm not certain why people are disregarding it when it comes to LLMs.

Sure. And we aren’t only using Claude. Nor is it essential to our operations. But it’s a tool we used widely and it’s gone (for the moment), and in a way that is untypical of most “vendor did unexpected thing that hurt our workflow” stories.

That's why I internally push to use Bedrock. If AWS flakes or bans the company account I've got bigger problems. Putting more eggs in the same basket is a counterintuitive solution to the problem of someone else holding my fate in their hands, but at least it's a platform that's been around > 5 years.

+1. Companies like AWS are not known for this kind of treatment. Even Google does this. The norm seems to be to randomly ban and ignore emails. It's a whitelist to build customer trust, not a blacklist of bad companies

> One cannot set up a business where another company becomes critical to your operations.

This describes a large percentage of successful businesses that exist today. Not even just in tech.

Doubtful. Do they use vendors? Sure. Perhaps even in critical functions. But that is not the same thing as a specific vendor being critical. There is almost always a business continuity plan, and frequently you'll find a full risk management plan with documented risks and mitigations. How far they take those planning efforts varies greatly, but I've never seen a medium or larger business that doesn't have at least some basic risk mitigations in place.

This is true, although different companies manage their vendor exposure with varying levels of effectiveness.

Often, it's ideal to use several / all of the vendors for each thing, and play them off against each other. e.g., have some of your database stuff on oracle, some on mssql, or some cloud stuff on aws and some on azure, make your apps portable, and tell them both that you'll switch to the other unless you get a good deal, with that being a plausible threat because you're already using the other one and know how to make your stuff work on both, and occasionally rotate apps between vendors, or change the mix from 50/50 to 60/40 just to show you can.

Of course, the vendors will be trying to work against this and will want to do some supposedly amazing deal if you go exclusively with them for everything... which might be attractive in the short term, but opens the client up to getting screwed in the long term once they fall into all the lock-in traps and lose the very _ability_ to switch vendors.

In Anthropics defense they make sure that no company can rely on them by being down all the time... And LLMs have become a commodity these days, so you can seamlessly just fail over to another supplier WHEN they go down.

> No, it isn't. No LLM platform ever will be. No platform or vendor of any kind ever will be, if we are being honest. One cannot set up a business where another company becomes critical to your operations.

Most of companies in the world have done that with Windows, though.

New attack vector just start asking chat bots questions that violate TOS and get the whole company banned.

So, what's new? I guess it used to be Paypal, then Google, and now it's Anthropic are randomly banning their customers.

For a multi-billion dollar megacorp, dealing/fixing things for individual customers is in the too-hard basket.

Same, just had a Max ($200) account that I had for the last 10 months banned without explanation, which is weird since all I use it for is to code landing pages and write emails.

Brutal. The silence is the worst part you can't fix what nobody will explain. Hope someone at Anthropic sees this and actually responds.

The same thing happened to us too. Fresh company account, banned right away. Two weeks later we get an automated email stating that our appeal was reviewed but they could do nothing about that. I can’t even begin to describe how ridiculous this is.

Correction: my old number of 70 users was outdated. We actually have 110

Same thing happened to us with OpenAI. Appeal did nothing.

Oh no! If you don’t mind, could you share some context on what you were using it for and at what scale?

I have used it for personal coding projects - expo react native projects and a Django backend. my account got banned, as the company owner non the less. My teammates were using it for the company projects (all coding through the codex app / vs code) and they were not banned.

Just guessing but IMO bans confirm the bubble. They subsidizing to gain customers while pro/max plan is reasonable, costs of tokens on their API is much higher. Moving it to on-prem might be future of AI.

It takes 3 weeks to get past the AI customer service bot, assuming you ever do. Good luck.

Reply on news.ycombinator.com