Hi everyone! We're on the To The Moon channel. Tech news, insights from Silicon Valley from all over the world, and everything about artificial intelligence. So, today I'll raise a fairly important and serious topic, which concerns the blocking of accounts and, first of all, of course, the blocking of Anthropic Claude and Claude Code accounts and the blocking of OpenAI ChatGPT accounts. Who gets their accounts blocked, what accounts get blocked for, why many people get their accounts blocked even though they worked for two or three years, say, and did it constantly.
Why this happens even when a person feels they shouldn't have been blocked, I'll tell you today. Plus I'll single out about 20 parameters, look, which really affect blocking and which you need to keep an eye on. My video is definitely not made for getting around these systems or somehow working with them better. But the topic is very important and necessary, including even for people who use ChatGPT or Claude Code in their regular work completely legally, live in a country where it's allowed, and aren't doing anything strange at all.
I already told you that literally in a recent episode my wife and I were driving back from Napa, and I was showing, using her as an example, that she runs a huge number of chats, yet doesn't really find her way around, doesn't fully understand in which situation she might have problems with ChatGPT. And the problem — well, some see the problem as a kind of threat, that someone will get access to your data and you'll have some issues because of it. But I still keep showing everyone the problem that you essentially don't fully realize what data the system analyzes.
By the way, I want to start with the fact that there's a feeling — I'm going to talk about these 20 points, but there's a certain feeling that if you've ticked the box "don't train on my data", the system essentially analyzes nothing. Let's go through it once more. So, when you tick the box "don't train on my data", that's a fairly important thing — you're not training the model on that data. And yet, for all that, the corporation does have access to that data. It says so openly.
Openly, chats or certain requests are analyzed by artificial intelligence selectively, or, well, if you enter some request connected with ill intent, for example, I don't know, committing some terrorist attack, right, then that request will immediately go for analysis by separate models and essentially into separate procedures of the companies — OpenAI, Anthropic, obviously Google too, and xAI. And they'll analyze it. If they see a threat further, they pass it to a person who sits in the office, in the company, in the security service.
If that person really sees a threat further, he can pass it to higher authorities, or he can even pass it to the police, right? And there are known cases where OpenAI handed data to the police. So, there are a great many cases of accounts being blocked now. Accounts are being blocked from different countries.
Here a person writes that he worked from an unsupported country, although he stated as the owner that he lived in a permitted one, and the block happened back in January 2024. So, further, by his account, support directly explained the block by an attempt to access from an unsupported country and the use of a proxy. The user admitted that he uses a proxy, but essentially nothing was changed for him. Here they write: "Russia, a long-standing account was deactivated, the appeal was rejected."
But the author reported that he lives in Russia, had used — in this case, in this example — ChatGPT for years to learn programming for projects, and connected through various VPNs. His account was deactivated and the decision was upheld. So, this, by the way, is also an interesting aspect: long use, payment and so on in no way, under no circumstances prevent a later deactivation. There's a case, for example, from Iran, absolutely fresh, the beginning of October, and support responded quickly.
A user from Iran reported that the account works on other devices, but the Windows app asked to confirm the phone and didn't accept the Iranian number, right, which starts with the code +98. So, support replied directly that you can't use this country. That is, the restriction was essentially triggered by a completely different criterion. In Belarus, too, a person writes that in May 2023, so, everything worked for him, then the account closed, and he generally used, at the beginning he used the service without a VPN.
Well, it's clear what it was like in those years. And then the ban came. Overall, if you look now, accounts are being blocked everywhere. And I want to give you the background. At Anthropic — well, basically the most widespread, mass thing is, of course, Anthropic. And people are incredibly interested in this not happening at Anthropic, because at Anthropic — well, with the help of Anthropic Claude — a huge number of systems are being created. So, what happened in the first place? On September 4, 2025, Anthropic announced restrictions for foreign companies whose ownership structure places them under the control of some unsupported jurisdictions, right?
And they even mentioned that ownership of more than 50% in such a company by that unfriendly jurisdiction — well, for example, the company is in Russia, say, or in Iran — they'll carry out this block. Next — I'll remind you once more that today we'll go through these 20 specific criteria for what people get blocked for. It's important to really understand them, and to be aware of what's going on in general. By the way, I want to note that there are very interesting cases, right, specifically described cases at Anthropic, where a person, for example, went to some country, travelled there, for example, on a business trip or, I don't know, to relatives, to his grandmother, it doesn't matter, came back, and a flag remained on him in the system.
And that flag — he can explain later that he went to see his grandmother, but that flag can later have a negative effect on the whole account history. This is a big question in general in the modern world — where to launch which accounts and when. We'll talk about it now. So, next. On May 12, 2026, Anthropic already described in more detail what a block can be for, including the user's location, and also the registration of the company's headquarters, ownership of the company, and, and, say, the location, for example, of its employees abroad.
By the way, I want to say that back in an episode last year, in the autumn, at about this time I was in Paris, and I remember very clearly how part of my services got blocked. I think Sora stopped working for video generation. It still existed back then. You know, right, that Sora, Sora no longer exists as a product at OpenAI. And we also discussed in our To The Moon episodes that in general ChatGPT works, as it were, less well from other countries. Tanya also talked about this, that from China ChatGPT seemed to start working differently. And I noticed very strongly that from Europe ChatGPT works differently.
Now this is already kind of becoming fairly normal. And it's clear that they don't just provide different services in different regions. For example, Dots still doesn't work in Europe — it's officially written that it isn't supposed to work — and, among other things, the model itself can work differently, answer differently, do things differently. And there was a very cool big study — I don't remember whether it was at Anthropic, or at OpenAI, or at some outside company altogether, done externally somehow — on the subject of each system in different countries being able to give different, politically oriented answers.
And politically oriented not on the basis of some kind of world notion of freedom, or, for example, a world notion of laws or politics, but, for example, this is how it goes in this country, and so this model would start restricting everything for that state, even though the model itself is closed in that country. That's certainly surprising. And why did I decide to film this episode? Because on October 8, 2026, Anthropic announced an update to its general set of rules, and this set of rules took effect on November 12.
And right now blocks are simply starting en masse, they're happening in a lot of places, and it's very interesting. And what's more, the company's regional section was directly, directly called a clarification of how the existing policy is applied. I'll explain now, I'll explain now what this is in general and how it works. So, in general, who do the restrictions basically affect? You should understand that there's the notion of a corporate plan, but there's also the notion of a personal plan, right?
And essentially Anthropic builds, shows all the restrictions through business. But then it says that if you've made your personal account for corporate needs, you'll also fall under the block. That is, you should understand this perfectly well. About these 20 parameters, we'll talk further below. So, what's the situation? A person uses, for example, Claude, for example, from Russia, right? Use is not allowed. Next.
An employee of an American company came to Russia on a business trip. Use while he's there is not allowed. For example, a company is registered in Iran, and its developer, for example, has left and works in Europe. That is, the developer's move — by the way, in Europe such a case is hardly possible. Let's put it this way: a company is registered in Belarus, its developer has moved to Serbia. So, the employee's move doesn't lift the restrictions on its staff. So, a company is registered, for example, in Kazakhstan, but belongs to a Russian parent company.
So then, foreign registration doesn't cancel the restriction on ownership or control. And here, of course, the question arises: how will they figure this out? Right? That is, it's clear that there's the form of ownership, which in theory can be looked up, but imagine — you'd have to look at every company that registers. And if you're an employee and registered your personal account, how will it figure out that it's this company, how will it figure out the legal entity? What's more, it's very hard to make sense at all of company names and legal entities, right?
Especially people in Russian-speaking spaces know this — that, for example, in Russia, say, or in Belarus one company can often have, say, three, four, five legal entities, or they'll be named differently, or use some other different forms of ownership, which is less common or works differently from the point of view of Europe or the US. But still. So, a company is registered abroad — there's this too — but its headquarters, for example, is in Russia. And the headquarters is also an independent ground for a restriction. So, these are examples for you of the published terms that Anthropic publishes.
So, there's an opinion floating around the internet that if a person has, for example, a Belarusian passport or a Russian passport, or an Iranian passport, then in any case there'll be control, there'll be a restriction. I want to note right away that this thing is very relative and essentially not real. And literally this morning a friend of mine gave me a complete case. He has a Belarusian passport, but he has residency in another country. And clearly he sent this data to Anthropic, and within that data everything went fine.
But he didn't send the Belarusian passport. And in general this story with passports can certainly raise questions. Here I recently had a case: on my account, on my YouTube account, they wouldn't give me one award because for some reason they considered the account to belong to Russia or Belarus. And that restriction was lifted only after I gave them, in detail, month by month, my residence, my movements. That is, my American documents didn't even mean anything to them. They had some other logic of their own, right? Even though they don't see any additional passports or details or anything else.
So, at Anthropic, clearly, absolutely all products are affected, and at OpenAI all their products are affected too. And, by the way, I want to remind you that in July of this year there was a topic that wasn't just about blocking in certain specific countries. In July of this year the question was discussed that possibly Fable — and that's the coolest model at Anthropic, there's this model of the Mythos family, and the coolest is Mythos. But, by the way, it wasn't in July, it was June 12. The US government demanded that access by foreign nationals to Fable 5 and Mythos 5 be restricted, right?
And here it's very important that foreign nationals, including those located on US territory. That is, it wasn't a question about residency at all. This question even included — it caused a huge problem in the market, in the discussion, because this topic even included people who, for example, a person works at Anthropic with a passport — it doesn't matter — of Canada or Mexico, or Italy, or Belarus. And he, for example, simply has a green card, and he too would have had to be blocked from these models.
So, I'll remind you that the models that came out later, Fable 5.1 and Fable 5.2, which has just come out... Watch my episode — I released one about 10 days ago about Fable 5.2 and why you should use it. Fable 5.2, which has come out — they're available to, well, a wide, wide range of people, they're not limited only to US citizens, and that's important to look at. Mythos, I'll remind you, was restricted and remains restricted. Why did Anthropic introduce this policy at all, and why did this policy begin?
They gave this explanation last year, exactly a year ago. And the main example of this policy was China. And Anthropic named several risks within it. The first was forced transfer of data to the state. That is, from the point of view of the company's position, a business under the control of certain jurisdictions may be obliged... Look, it's not a question of a state-owned company — a business is obliged to cooperate with the intelligence services, regardless of where its foreign structure is located, right?
Clearly their second, very big block was military intelligence use and the possibility of using Claude Code, well, in this case ChatGPT, through various foreign subsidiaries to develop the corresponding systems. And, obviously, let's be honest, well, clearly bad people in the world try to create various anti-security systems, to create various problems for the world, for people, for honest people, honest citizens and so on. Well, plus all countries have different laws, right? And what's considered absolutely illegal in one will be super legal in another.
So, a big story, a huge one, has been the training of competing models on the answers of Claude, Claude — well, in this case ChatGPT — everyone everywhere fights this. But Claude is the most widespread issue among those declared. And Anthropic has repeatedly published a huge number of cases where a large number of accounts were created. In reality it's millions of accounts, of course, being created to distill data. We've talked about this parameter many times. That's when third-party models are trained on Anthropic's or ChatGPT's own models.
And there's direct research, a direct indication, that in China you can buy API access to Anthropic, often at a huge discount, not just 20% off, but sometimes five times cheaper, because this, this API is run through a special middle layer that distills everything completely, right? And you get your own result, right? That is, you do what you want, but before that another company takes all your data and trains its own models. And essentially, if there were no possibility to distill Anthropic and OpenAI, then I want to tell you this: Chinese models would most likely not exist at all.
That is, if today the American market still has a competitor — China — then if there were no distillation and no contact at all of foreign employees with this, that would have been a big problem. We'll talk more about this topic regarding employees. So, well, and the fourth task, why Anthropic introduced such a policy, is because it was protection of US strategic interests. They also mention the US and its allies. And Anthropic formulates this motive very directly. What's more, Anthropic formulates very directly that it's prohibited to use Anthropic's models against US citizens or against US residents and against US residents.
At the same time, from the point of view of use in the world, for example, the US government or the US military government using Anthropic against citizens of other countries is not prohibited. Although with ChatGPT, OpenAI, this was stated very clearly, OpenAI stated that it's not prohibited. I'll remind you, there was a whole scandal, right, with Venezuela. It's still going on, by the way, this very big scandal, we don't know how it will end, because now the Democratic parties have come to a fairly strong imbalance.
That is, they got power in the US — they'll get it, rather, from January of next year. Remember, they haven't got it yet, they'll definitely get it from January. And that will certainly bring a huge number of questions, including into how all these systems work. In general, in the first quarter of next year we'll see what the real policy towards all these systems will be and, well, what will happen, right? Because it's, well, not a simple story.
So, before we move on to these 20 points, I want to say this: I've said many times that cool companies such as Anthropic and OpenAI will definitely introduce restrictions. These restrictions have already begun, right, they already began when systems such as Mythos were given only to a select few. There are special systems for biology, physics, mathematics inside OpenAI and Anthropic, which are also, also used only by designated people. There are special systems that are used only for states.
And it's clear that if Italy buys itself an OpenAI artificial intelligence model, or the UAE — right, they cooperate very closely with OpenAI — then, obviously, that model won't be of the level and class that will be used in America. I'll remind you that the use of a model depends very much on what infrastructure it runs on. And right now there's a huge problem in the infrastructure market too. Not only has the market for artificial intelligence technologies and computing been shaken financially very hard, there's also this problem that happened just now with Jensen Huang, and there's an opinion that in the coming days a huge investigation will come out about NVIDIA, about what was going on there. That's a big question.
You can see it's becoming very strong politics, right? And that's why, of course, I'm sure that Anthropic and OpenAI will at some point verify the identity of absolutely all people. It's strange that they haven't done it everywhere right now. And they say there are two opinions on this. On the one hand, they can train on data from, of course, other countries, and also let the world's cool developers use their models. Because if they restrict them completely, while America is closed for entry, as it is closed now, then America will lose an incredible number — these companies will lose an incredible number of truly cool people.
That is, probably — today I spoke with my partner in IT, he said this will work if America opens entry to all these people and the block happens at the same time. We'll see. So, point by point. Write right now. Now I'll start telling you about these 20 points. Please write right now what you've run into, who has what problem, what problems your friends have. Let's gather a big pool like that and see what happened where.
So, point by point. I'll start listing these points for you, the important ones. You can write them down, you can note them somewhere. Later we'll post the transcript on the website, and you'll be able to look at it as a whole, and I'll give some comments of my own. Be sure to read it. Incredibly important, necessary, specific information for understanding what's going on. The first thing is the country of connection. So, the country and roughly the city are determined by the IP address, say, and the network address of your connection. And Anthropic directly confirms this check.
So, point number two is the time zone. So, it matches — this, by the way, is probably the most important story of all, these first two points. Although all the points are important, right? We'll talk about it now. So, does the browser's time zone match the presumed location? It's only about the zone, not about the clock being a few minutes behind, right? And notice, look, that if you change this data, it's all saved on their side. The whole history of changes is saved, right?
Here on my website there's an automated system, it was written by artificial intelligence, but it isn't artificial intelligence. Right away, when it was writing, for example, for the besolid.com website, the system made it so that it automatically detects unauthorized logins to the site. That if one user got access to some paid program — and I have a certain training program on a person's personal development, on a person's spiritual development, on people's spiritual development, right?
That if one person logs in from different addresses or from different VPNs, from different time zones, it highlights these people directly and says: "These people may be passing the account on to someone else." So, the third point is the connection type. Are you connecting via a home network, do you have a mobile operator, a data center server, are you using a VPN, for example, a virtual private network, are you using a proxy as an intermediate server, or are you using, for example, Tor, say, an anonymization network. This is also an important story, by the way, right?
Because if I, for example, right now I'm in Mendocino, that's four hours' drive from my home, where I live in Silicon Valley, in Los Altos Hills, about two, roughly three hours from San Francisco, and on the ocean. And here I've logged in, here the network is the Xfinity provider. That is, they see that I connected directly to Xfinity, right? Or I'll connect — I don't know, I have a mobile Starlink lying around, I'll connect to it, and they'll see that I connected to mobile Starlink.
And I can only connect to mobile Starlink directly. That is, it's clear what's going on with me in terms of connectivity. Next there's the fourth point — the reputation of the address. Is the address linked to past attacks, spam and other abuse? And I want to note that this is a big question, right? If you use, for example, corporate addresses, someone got something blocked, or you arrived somewhere in some country and connected to some address, or you bought a VPN, you don't know what this VPN is, what address it has, how it's set up, and if it's not a system you made yourself, right?
So, and even if you've arrived in another country — here I sometimes come, I don't know, to Europe, and I need to open a VPN to connect to certain American services, to log in, right? Sometimes to visit some American store or some other one, right? The system can diagnose it, accidentally see that I used a VPN, and mark some tick for itself, set it. So, the fifth point, fifth from the point of view of the algorithm, is movement history. If you change countries often, move unusually fast, log in from distant countries within a short time, right?
Again, right, if you have, for example, a Singapore time zone, and you suddenly logged in from Scotland, then a question arises, right: why, what is this you've got, what happened, right? And then suddenly Italy, I don't know, switched on for you. The sixth point is language and regional settings. So, the browser's preferred language and other environment settings, and the available data. The Russian language by itself doesn't prove you're in Russia, right, you should understand that. But I want to mention one nuance. I develop and create a lot of everything in Russian.
That is, my English is quite fluent. But I still digest complex texts and large volumes of information in Russian. And very often — Anthropic links me very strongly with Russia. That is, not with Belarus at all. I, I, so, was born and raised in Belarus. I really did live in Russia, but I'm a citizen of Belarus, not a citizen of Russia, I never have been, right? And the interesting story is that it links me not with Belarus, it links me with Russia. Why? Because for the system the Russian language is still, is, is Russia, it's automatically the Russian market. And if, I don't know, on our To The Moon channel 30% of people are Russian traffic.
Notice, by the way, YouTube is blocked in Russia, but at the same time it determines... That's interesting, right, Dim? At the same time it determines 30% of the people on my channel, or however many, or, say, 20% of people — it determines that they're in Russia. How does it do that if people from Russia can't log in? The same with Instagram, right? You know, right, that if you click, click in Instagram on Russian bloggers, on their location, you'll see that their location will say Russia.
Although in fact, although in fact, how can the system, in general, know that it's Russia? After all, people from Russia can't, can't — it's not allowed — log into Instagram, right? It's impossible to do in any way. And just in parallel I was opening it right now, to look at how much Russia I have. Actually, by the way, over the last 28 days it's not 30% for Russia there, it depends, it's even less, even less. And roughly at the same level right now there's Ukraine, Kazakhstan, Belarus, Russia, right?
Then Armenia starts. So, well, the US, by the way, also gives a decent volume. And the next story is point number seven — the "device fingerprint", fingerprint in quotation marks, right? So, it's the combination of browser, system, screen, fonts and other characteristics that makes it possible to recognize a device between visits. Look how much deeper it's already gone, of course. And someone here will even say: "What is this? How do I change the settings? How do I deal with it?" And that's a serious problem further on.
That's exactly why Instagram or YouTube does this with ease. Next there's the notion of links between accounts. For example, a match of device and stored identifiers. For example, many accounts linked to one visitor, right? It can be a common story, especially if you use, for example, several accounts, log in, log out. By the way, I'll remind you that neither OpenAI nor Anthropic prohibits creating several accounts, but they limit their number. I think ChatGPT allows three, if I'm not mistaken.
Anthropic officially allows opening a personal account and a corporate account. So, the ninth point is contradictions in technical data. For example, the browser reports one operating system, while the characteristics of the network connection point to another. That is, it says that somewhere you're falsifying information for systems such as Claude Code or ChatGPT. And, of course, the system wants to figure out what's going on there. And I'll remind you how it figures things out now. It immediately requests documents from you, photographs you, wants to check you completely.
The 11th point is automation and unusual activity — the frequency and sequence of requests, mass registrations, session characteristics, signs of programmatic browser control. That is, it's a more fashionable, new condition. We all understand perfectly well that Anthropic is on top of this, right, just like OpenAI, because various agents are built on their base that try to trick browsers, right? They know all this completely. The 12th point is payment geography: the card's country of issue, the billing address, the country of connection.
Their mismatch can be taken into account by payment protection, right? And this, this, of course, the topic in general, the topic of payments, is critically important, actually, and it's not simple. Today many countries in which — here, in particular, Anthropic is blocked in Russia — people try to pay with the help of third-party services, but they don't fully understand what these services are in general and how the payment is used, how it goes through, right? And what exactly a company such as Claude Code will see in the end. The 13th point is payment history. There's payment history too, look.
Here's what's connected with payments, besides, say, the card's country of issue: there's the notion of failed payments, declined transactions, chargebacks through the bank. That is, if you have this, it can affect the understanding of who you are, where you are and what you're doing. The 14th point is account data: name, email, phone, the addresses given, and the consistency of this information with the rest of the data. Anthropic reports collecting such information. And in the example I told you about above, about Iran, the system immediately says, for example, that the phone number isn't suitable. That is, a huge number of details.
I'll note, you gave an email, for example, with Gmail, and Gmail knows where you are. They have their own tracking system. That is, you, for example, don't protect your email through any tools — many people don't protect it, right? Or you use it somewhere, and the email shows up in another place. And here's the account that exists. And the 15th point is identity verification. So, if there's a document, a photo of the person, the result of the check and whether it matches the account owner's data.
And, accordingly, here's what Anthropic does: it runs such additional checks, right? So, the 16th point is the company and its owners. Only look, we saw that they block by company, but the first 15 most important points are about something else entirely, right? The company — what registration the company has, where its headquarters is, what the ownership structure is, physical control. That is, this relates quite directly to Anthropic's regional restrictions. The 17th point is device geolocation, the coordinates provided by the system.
In the browser this requires the user's permission, including one granted earlier. But don't forget that often somewhere someone clicked on another device, or pressed something in the browser, or pressed it in the system. There are a great many details. By the way, ChatGPT now has a feature for activating geolocation detection. I don't know whether Claude has it, Claude. If it does, write that it has it. Why is this feature useful in ChatGPT? Because ChatGPT, knowing your current geolocation, works better in terms of chats and Dots. In the comments, please write which points concern you, right?
We'll continue now. Write which points apply to you, what you run into, who has had blocks. Next, the 18th point. Additional network information. For example, WebRTC is a browser communication technology, and under certain conditions it reveals additional network addresses. It doesn't guarantee that the real address is determined, for example, behind a VPN, but still there's the notion of additional network information. The 19th story. The content of requests and files, right?
That's the information passed to the service and signs of rule violations. Anthropic allows the use of content to monitor compliance with its terms. And we've talked about this, right? And here's the 20th point. It turned out that there are 19 points here. The 20th point — I want, want to add a few more such additional particular aspects. Well, first, imagine you logged in from a computer, or a person logged in from a computer in some isolated environment. He needs to do it this way for the company, so that there are no problems and so on. Even if he uses it from a permitted country, he wants to minimize his risks.
But then this same person accidentally logged in from a phone once somewhere, and that's it. There are always absolutely these additional peculiarities. And, of course, I'd like to extend the 19th point into the 20th: today we use systems such as ChatGPT, Claude Code so widely in terms of information that at such a volume it's, of course, impossible to truly hide your location and details, right? And it's one thing — I drive, I move around and ask: "What's here, which café is close by? Where is the nearest charging station for electric cars here, in this village?
On which beach can you walk with a dog?" That is, you ask a great many local questions. Or: "What's going on at school?" Or, for example, I have four children going to school in the valley. Well, clearly in Silicon Valley, where I live — two in Stanford and two in Palo Alto. And it turns out the system reads their emails every day, it sees that we've received these emails, right? Plus Codex, for example, has a lot of access to my computer, or, say, to my home server. Well, that is, it clearly sees where I am and what I'm doing.
But there's a nuance now for everyone in the world. Why do you need to be careful? You, for example — I have a project, I do a lot with the Russian language. And Russian is 250 million or so people in the world. And half, for example, live in Russia, half live in other countries of the world. But the system, as I already said, Claude, can diagnose that the Russian language matters in Russia, and set some ticks somewhere, right? Even though you say: "Listen, there are Russian people everywhere."
What's more, you're not breaking any laws when you provide services, ordinary services to people — for example, watching the To The Moon channel, right, or listening to a program on spiritual or personal development. But these systems may have some practices of their own, some nuances of their own, and may flag you if you asked something that should have been asked, for example, differently. Here I, preparing for today's episode, asked a lot of questions. And I'm thinking — I'm asking about access and about security.
Won't it happen that ChatGPT thinks I'm trying to hack something? I tell it: "I'm doing this in order to film an episode about it," right? As if trying to sort it out in advance. Many people, by the way, try to get around the services, and for me that's an absolutely strange story. I don't support it, and I'm sure it doesn't work, right? That's using certain middle layers, when a certain system is used that resells via API, for example, the Claude API or, say, the ChatGPT API.
But you should understand that Anthropic has a single rule there: it obliges these companies that resell access to also be sure to keep track of location. For example, Salesforce has its own location detection. If you're inside Salesforce, inside this huge infrastructure for managing sales and marketing and all of a company's commerce, and use Claude via the API, then if you've moved to some other country, it will temporarily switch off the Claude API for you, but offer something else, right?
So here, too, you need to be careful in terms of various movements and of what may, what may really happen with it. Be sure to share your comments. A super important topic. Forward it to other people, save this episode for yourself, because you'll need it for a long time yet. And don't forget to support our channel, give it likes. We have a lot of new episodes coming out. This week there was an episode on Monday about Cybercab. Let me remind you, Cybercab is not Cybertruck. There, it's a car that has no steering wheel. Take a look at it, it has already appeared on sale. And come to our channel.
Yesterday we had an episode with a new author as well. He told about four very interesting, great cases there, about quite lively cases. I even remembered how in my youth I played, played Counter-Strike, and was glad about what things can be done in games now. Although, of course, I don't play them. Bye, everyone!