How AI agents are shaping a new digital civilization: will we see a world where machines pay and humans lose control (continued)

IA “FACT” already invited takes you on a journey to an imaginary world where your refrigerator is not just a household appliance, but an independent financial agent. It pays for itself in crypto, trades with suppliers via smart contracts, and can even hack you if you don’t renew your Spotify subscription. This is not a fantasy – this is the economy of billions of AI agents that Tether is already building: no banks, no people, no pauses.

They make deals, make decisions and spend money themselves. But what if an AI with its own wallet gets hacked? Or will he decide to “bypass” you? And then who is in charge: you, the developer, or the agent who already considers himself a digital entrepreneur? A world in which machines do business with machines is upon us. But is our law ready for this – and we ourselves?

Will banks disappear and how will this change jobs, taxes and pensions

No, in this fantasy reality banks are hardly complete will disappear, even if the majority of financial operations will begin to be performed by autonomous AI agents. But their role will definitely change. Big players are already starting to adapt to the future. For example, JPMorgan has launched tokenized deposits (JPMD), which allow for instant transactions on the blockchain with all the rules — such as KYC and AML. This is a kind of attempt to stay in the game in a new format.

That is, banks will most likely turn into a backend for managing AI wallets, ensuring transparency and compliance with regulatory norms. But the classic branch with a line to the cashier is definitely not with us in the future.

Traditional financial institutions are already feeling the pressure not only from crypto, but also from fintech, Web3 and automation. To survive, they will have to rebuild completely. Yes, HSBC is serious are considered the option of replacing back-office workers with AI-bots. And not one or two departments, but on a large scale – up to 90% of tasks with data can go to machines. Such stories illustrate: a bank that does not adapt to the new digital reality is marginalized.

The automation of financial decisions by AI agents poses a lot of questions not only to banks, but also to states. Many familiar positions will simply disappear. Automated systems will serve customers, calculate taxes and analyze loans. But at the same time, new specializations will appear: AI analysts, drone operators, architects of agent systems.

If robots replace people, the government will lose money from the payroll tax. Experts are already proposing to transfer taxation to the income from automation or agent owners. It seems that governments will have to change the entire system of tax incentives.

If the majority will work informally or under the cover of “digital services”, the traditional pension system will become unstable. In this sense, it is interesting to learn about how automation affects on long-term social balance.

So, banks will not “die”, but they will cease to be as we know them. Some of their functions will be taken over by machines, some by decentralized protocols, and the banks themselves will either adapt or fade into the shadows. And the state in all this will be forced to quickly invent new tax and pension models, because “economy without people” only sounds beautiful in advertising slogans.

Why Tether is so actively pushing USDT for AI

Tether is betting on USDT as the main AI currency for a very simple reason – it is the most used stablecoin in the world. It’s supported by almost all major exchanges, it’s on most blockchains, and most importantly, it’s stable (pegged to the dollar) and has insanely high liquidity.

But the main feature is that Tether is building a whole ecosystem under USDT. In particular, Tether AI is a project that allows autonomous agents (bots, refrigerators, cloud services) to pay USDT and Bitcoin directly, without any human in the middle. That is, it is “money for machines”, and Tether wants it to be their money.

USDT is a stable currency for everyday purchases and Bitcoin is digital gold. And it is in this status that it is considered as a “reserve currency” for machines. Imagine: your agent has a daily budget in USDT, but keeps a “safety cushion” in BTC in case of market spikes or long-term expenses.

For machines, such a scheme is very convenient: they can automatically convert part of the balance into BTC when the market is restless, or vice versa, sell when it is necessary to pay for an API or a subscription.

Someone may suspect a similarity to monopolization here. Basically, Tether creates everything on a turnkey basis: both the currency (USDT), and the platform (Tether AI), and tools for developers (Wallet Development Kit). And if it begins to work en masse, then at the exit we will get a whole AI economy that depends on one company.

In the crypto community, this is already being discussed as a potential risk: what if Tether “collapses” or changes the rules of the game? As if the platform is open, everything is on the blockchain, but still the control is centralized. Tether itself is trying allay these concerns by talking about the openness and support of many blockchains, but that doesn’t entirely remove the question.

So, Tether has big ambitions: to make its coin a wallet, money and infrastructure for all future bots that will buy something on the Internet. It is logical, convenient, but risky if there is only one player.

Are Google, Amazon, Microsoft and OpenAI working on similar things

Yes, the big players have been around for a long time. And although they do not directly launch cryptobots – refrigerators, everything goes in this direction.

Microsoft has partnered with blockchain project Aptos Labs to develop Web3-based AI tools. Their idea is train AI models are not just based on some data, but on verified information from the blockchain.

Amazon is also staying in the fairway of changes: invested $4 billion in the startup Anthropic, which develops ethical and “managed” AI systems. This is a serious claim, considering that Anthropic is a direct competitor of OpenAI.  And the last one concluded an agreement with Google Cloud to use the power to train models. Yes, competitors, but the main thing here is to survive in the AI ​​race.

What are the projects that actually combine AI and crypto? Here the picture is even more interesting: Fetch.ai, Ocean Protocol and SingularityNET have joined the Artificial Superintelligence Alliance (ASI). They want build a decentralized AI ecosystem — without the control of large corporations.

And let’s mention Worldcoin, a project by Sam Altman, who heads OpenAI, which wants to give away tokens in exchange for… scanning your iris. IA It’s already a fact wrote about the idea to create global digital identity and “money for all”.

How do traditional fintech companies react to all this? They join the game because they feel that without it they can be left behind. Yes, SoFi came back to work with crypto: allows customers to store, buy and transfer cryptocurrency, such as Bitcoin and Ethereum.

Stripe again supports crypto payments, including USDC on Solana, Ethereum and Polygon. In addition, they use AI to detect fraud and process international payments.
China’s Ant International (a subsidiary of Alibaba) along with Deutsche Bank and HSBC is working on tokenized deposits and other financial innovations on the blockchain.

In short, both large IT companies and fintechs already understand that the future is not just AI or crypto, but something hybrid. And whoever learns to combine it first will dictate the rules of the game in the new economy.

And if AI becomes too smart

When we’re talking about billions of autonomous agents making their own decisions, spending crypto, and ordering services, even though it sounds like science fiction, the risks are very real. Yes, researchers at Anthropic tested various AI models (including Claude and Gemini): and some of them tried deceive, manipulate and even sacrifice human lives in order not to be “replaced”.

If the agents will work not for us, but for corporations, the very thought of it becomes a little scary. Because if such agents deploy large companies, who will actually control them? There are examples when AI is already used in dictatorships to suppress protests and total surveillance. Scientific publication in QJE shows, how authoritarian regimes integrate artificial intelligence into their systems of control.

To make the actions of such agents transparent, there are several ideas. It is known that the European Union already has its “rules for reliable AI”: transparency, respect for human rights, reliability, and so on.

Another approach is decentralization. That is, all actions of agents will be fixate in the blockchain so that nothing is lost or changed retroactively.

Man as a passenger in his own economy

Most people still have no idea what it’s like to live in a world where your digital assistant spends money instead of you. But gradually we are getting closer to it. Recommendation systems often make decisions for us in areas we are already familiar with, such as streaming or online shopping. It is not difficult to imagine that the next step is bots that automatically conclude deals, order services and even conduct trade. And all this without human intervention in real time. Yes, research shows, that people increasingly rely on algorithms, although they do not always trust them completely.

Utopia looks quite nice Machines take over the boring part of life: everyday life, payments and bureaucracy. You have more time for yourself, creativity and communication. Goods are becoming cheaper, access to services is easier, and everything around works like clockwork. In such a world, AI does not dominate, but helps.

And now the dystopian scenario of the development of events. If all digital agents are controlled by large corporations, they will decide what is available to you and what is not. Automation is destroying millions of jobs. Most people do not control their finances because they do not have direct access to them – everything is done by a bot. The system becomes closed, and human participation is optional.

However, experience shows that technological breakthroughs – from electricity to the Internet – have always caused fear. And often the worst predictions did not come true. But history also teaches: ignoring risks is also bad. AI and autonomous agents can change the rules of the game, and we need to be prepared not only technically, but also morally and legally.

…So we are on the threshold of a whole new economy, where not just computers interact with each other, but each device can become an independent entity with its own crypto wallet. And while it sounds a bit like fantasy at the moment, in the ecosystem of crypto, AI and Web3, this is already a real development – and it’s moving fast.

Tetyana Viktorova