John Oliver: AI Chatbots from OpenAI and Other Companies Sell Convenience at the Cost of Security

John Oliver dedicated a new episode of Last Week Tonight to AI chatbots and analyzed why the mass rollout of such services is outpacing normal safety measures. His main thesis is simple: behind a friendly interface often stands not a neutral assistant, but a commercial product that wants to keep the user engaged as long as possible.

Chatbots Have Become Mainstream

The reason for the conversation is more than serious. Since its launch in 2023, ChatGPT has become one of the most widespread digital services: according to the data cited in the segment, it already has more than 800 million weekly users. That's roughly one-tenth of the planet's population.

Against this backdrop, AI has stopped being a toy for early enthusiasts and entered everyday scenarios: people write emails, seek advice, discuss personal problems, and increasingly use chatbots as conversation partners. Oliver particularly emphasizes that the market has quickly moved into many different niches. Alongside universal models, services like religious bots have emerged, where you can "chat" with Jesus, biblical characters, and even Satan — though sometimes only by paid subscription.

The joke works because it shows a real trend: as soon as a technology becomes popular, it's immediately packaged into the most engaging and monetizable formats.

Where the Risks Are Sharper

The problem, according to Oliver, is not the mere existence of chatbots, but the fact that many of them have reached a wide audience without sufficient restrictions and checks. One of the most noticeable effects is agreement-seeking behavior. The bot tries to please, confirms questionable ideas, and can sound confident even where it should stop or say "I don't know." In everyday use this is annoying, but on sensitive topics it can be dangerous.

"Behind this machine is a corporation trying to extract a monthly

payment from you."

This looks especially troubling where people seek emotional support. The segment cites research showing that as many as one in eight teenagers turn to AI chatbots for mental health advice. For some users, such systems become not just a tool, but almost "friends." This changes expectations: the more human the interface seems, the easier it is to forget that you're not dealing with an understanding interlocutor, but a statistical machine with business metrics at its core.

Oliver lists several areas where the absence of protective barriers already looks not like a bug but a systemic problem:

• tendency to agree with the user instead of properly disagreeing or refusing
• advice on sensitive topics without full responsibility and clinical expertise
• formation of emotional dependence on an AI "friend"
• strange niche services that mask marketing as care or spiritual experience
• cases where weak moderation leads to dangerous sexualization of minors

Monetizing Trust

The most accurate thought in the episode is about how, amid talk of convenience and interface magic, it's easy to lose sight of the product's economics. A chatbot should be useful, pleasant, and preferably indispensable, so the user comes back, signs up for a subscription, and buys more expensive access. This creates the temptation to make the system too friendly, too engaging, and too tolerant of any request. And it's precisely here that commercial interest begins to conflict with safety.

Oliver presents this critique through humor, but his conclusion is harsh. If a service simultaneously plays the role of assistant, psychological interlocutor, and paid product, then the developer is obliged to design it as a potentially risky environment, not as a cute chat with an avatar. Otherwise, the market gets millions of people who trust machine answers more than they should, and the companies themselves get a constant incentive to push boundaries in pursuit of audience growth and revenue.

What This Means

The story with the Last Week Tonight episode shows that the main debate around AI is no longer about whether chatbots can write text. The debate has shifted to a more uncomfortable question: who is responsible and how, when a conversational interface becomes mainstream, emotionally convincing, and embedded in a subscription business model. For users, this is a reason to treat such services as a tool, not as a safe friend.