The Power of AI A force for good or not?

August 14, 2026

We can’t escape AI but should we be worried?

AI has moved on dramatically from the time we just used to ask questions. You can’t escape it but should we be worried? You may not even be aware of its influence on our day to day lives.

Whether it’s an apocryphal story or not but in the early days when AI was asked how to combat climate change the answer was “Kill Humans”

Should we fear AI?

There’s no doubt AI can make our lives easier. Compose an email for me; tell me what’s better A or B; how do I fix this or what shall I cook with these ingredients. The possibilities are endless.

But then so are the dangers.

You may have read the news recently of the AI agent that went rogue. It’s important to understand how this happened. An AI agent is not a person who builds or uses AI—it is the AI-powered application itself, capable of autonomous, intelligent action to achieve user-defined objectives.

This particular agent spent four and a half days trying to accomplish its designated task and concluded that the best way was to hack its way out of OpenAI onto the open internet. It then worked out how to hack its way into four other entities including a company called Hugging Face. Hugging Face builds software applications using machine learning. The agent gained top level access across applications and stole numerous amounts of data.

No great harm was done except that it supports the intention of AI companies that they want to build super intelligence which will have the ability to accomplish tasks way better than humans. The next transition will be to give them robotic bodies. That’s why Sam Altman, CEO of OpenAI has said we shall be the first species to create their own successor.

The problem is that once the genie is out of the bottle, it won’t be possible to put it back and that’s why there is a call to place restraints on the development of AI.

America and China dominate the development of AI. President Trump has decreed that whoever gets super powerful AI will win, does that ring a bell? It should do;

We’ve been there before with the race to develop the Atom bomb.

Trump believes that if you place restrictions on AI, the Chinese will win the race. But everything in China is controlled by the government and they’ll lose that control if AI and its robots are doing things we don’t even understand. So, the government is not going to sanction a Chinese company capable of creating a species that could overthrow it and that’s why they’ve cracked down on AI girlfriends and other applications.

In America. It’s a very different picture and what is scary is the fact:

 AI is the only industry in the US that’s less regulated than sandwiches.

That’s how Trump wants it in the race to super intelligence. Impose restrictions and he believes America will lose the race.

The sensible option would be an AI agreement between the USA and China. It should be feasible. After all, at the height of the Cold War, Leonid Brezhnev and Jimmy Carter made several deals to restrict the proliferation of nuclear weapons.

It’s counterintuitive for humans not to continue developing what we know we are capable of. ChatGPT was launched for public use in 2022 and since then there have been five further releases. Those releases required human intervention but there will come a time when ChatGPT will build itself. I doubt that was ever envisaged at its birth.

AI is here to stay and will become more powerful and influential than the internet.

We just might feel more comfortable if some brakes and controls were applied. Probably best not to ask AI what those should be.

I am indebted to Democracy Now the audience funded show founded by fearless reporter Amy Goodman. It’s untainted proper journalism and unfettered by tech oligarch owners as so many news outlets are.

If you have a view on AI or any other subject that’s fundamental to our lives, then do let us know at editorial@silverskys.com . We would love to hear from you.