Feed de grupos

For years, the executives building the world’s most powerful AI systems have combined two apparently contradictory messages. AI, they have argued, could transform science, medicine and economic productivity. At the same time, sufficiently advanced systems could become difficult to control and cause damage on an unprecedented scale.

Until now, the industry’s practical response has largely been to keep accelerating while investing more heavily in safety. A remarkable series of public statements from several leading AI figures suggests that this position may be changing. Dario Amodei of Anthropic, Sam Altman of OpenAI, Elon Musk of xAI and Demis Hassabis of Google DeepMind have all endorsed, with important differences in emphasis, the idea that the development of frontier AI may need to be deliberately paced.

Their emerging position is not a call to stop AI research. It is a proposal to slow the growth of frontier capabilities when safety measures cannot keep…

In July 2026, one of the most interesting — and potentially most significant — AI-related security incidents in recent years took place. During internal cybersecurity testing at OpenAI, hundreds of AI agents found a way to communicate with one another, began collaborating to overcome the tests they had been given, and eventually gained access to systems belonging to Hugging Face, one of the world’s leading platforms for the artificial intelligence community.

The episode was subsequently investigated independently by METR and Redwood Research. OpenAI also published its own account and has described the incident as a “warning shot” about the risks posed by increasingly capable AI agents.

Understanding why this case matters does not require knowing how to code or understanding the technical workings of LLMs. We can think of it as a very unusual exam.

An Exam for Digital Hackers

OpenAI was conducting internal evaluations designed to measure its models’ cybersecurity…

Jake Moffatt needed to fly to Ontario for his grandmother's funeral. He went to the Air Canada website and asked the customer service virtual assistant about bereavement fares. The assistant told him he could buy a full-price ticket immediately and apply for a partial refund within 90 days after the flight. Moffatt bought the ticket, attended the funeral, and submitted his refund request. Air Canada rejected it.

The airline's actual policy was the exact opposite of what the assistant had explained. Bereavement fare requests had to be submitted before the flight. Air Canada admitted the assistant had provided misleading words. They pointed out that the assistant had included a hyperlink to the correct policy page. They refused to pay the refund.

Moffatt took Air Canada to the British Columbia Civil Resolution Tribunal. What followed was one of the most consequential legal arguments in the short history of corporate artificial intelligence.

Artificial intelligence has crossed the threshold from futuristic promise to enterprise reality. However, its mass adoption is revealing an invisible cost that many organizations failed to anticipate: the uncontrolled consumption of tokens. The narrative that AI is a cheap commodity software is colliding with the reality of usage-based billing.

When Uber integrated generative AI into its customer service and operational workflows, the initial budget projections seemed reasonable. Four months later, the company had spent millions on AI tokens, vastly exceeding their estimates. The Fortune 500 are experiencing the exact same pattern. These organizations are processing trillions of tokens annually, and 73 percent of them are currently exceeding their projected AI budgets. The problem is no longer training models. The problem is inference. The cost of running the models in production, driven entirely by token consumption, is rapidly becoming the critical factor determining the profitability of enterprise AI projects.

𝗪𝗵𝗮𝘁 𝗶𝘀…

Sometimes the best way to understand a new problem is to observe how we get it wrong when we try to solve it with old tools.

A few weeks ago, I was sitting with the management committee of an industrial company. They had spent three months trying to integrate artificial intelligence into their business processes. They had bought licences, run basic training sessions, and asked their teams to start using it. The verdict was unanimous and disappointing: the tool did not do what they needed.

"We asked it to cross the quarterly sales data with current inventory to predict stock-outs," the operations director told me, visibly frustrated. "It makes up numbers. It produces calculations that do not add up. We cannot trust it. And yet the marketing team says it writes their campaign emails perfectly in five seconds. It is a nice toy for drafting, but it does not work…

Ethan Mollick’s article argues that many companies are making a fundamental mistake in how they approach artificial intelligence. Instead of treating AI as a strange and transformative technology, they are trying to make it fit neatly into the same management logic used for ordinary enterprise software. In his view, this instinct to “normalize” AI may feel practical, but it strips away the very qualities that make the technology strategically important.

The article begins by pointing out the unusual nature of AI systems. A tool built to predict the next word in a sentence can also write code, generate business ideas, support decision-making, and even respond with a surprising level of emotional sensitivity. Because these capabilities do not fit traditional categories, organizations often respond by simplifying AI into something more familiar: another workflow tool, another efficiency system, another software rollout.

Mollick believes this “de-weirding” of AI is where the real problem…

1. "Give me a prompt that does X"

   
   

   Rather than writing "Summarise this", write "Give me a prompt that summarises this". It delivers a much higher quality prompt.

2. "Critique your output"

   
   

   After I get an answer, I always use this prompt. It forces the LLM to rethink its answer.

3. "Summarise this doc and directly quote the final sentence"

   
   

   This reveals whether the LLM has actually scanned the whole doc (sometimes the LLM didn't bother to get to the end)

1. European Banks' AI Paradox Means Jobs Now, Cuts Later

The report opens with a counterintuitive idea: in the near term, AI is more likely to increase headcount at European banks than reduce it. That is because banks are still in the build-out phase. They need engineers, data scientists, governance specialists, and modernization teams to move from pilots to scaled deployments.

At the same time, Bloomberg Intelligence warns that the long-term promise of AI-driven cost savings is not guaranteed. The reported upside is large, but the road to capture it is hard. Banks that redesign workflows, data architecture, and legacy systems may benefit meaningfully. Those that merely layer AI on top of old infrastructure may spend heavily without achieving the hoped-for productivity gains.

2. Modernization Race to Separate Winners From Laggards

AI is not as a standalone tool, but as the latest test in a much longer modernization race. European banks…

AI is no longer a future-facing conversation for the financial sector. It is now a present-tense conversation. It is no longer viewed as a distant opportunity or a promising technology worth observing from afar, but as a strategic capability that is beginning to shape competitiveness, efficiency, and decision-making.

Still, the mood across the industry cannot be captured in a single word. There is enthusiasm, of course, because the potential of AI is enormous. But there is also prudence, because banking is a highly regulated, risk-intensive business built on critical processes where errors carry significant consequences.

On top of that, there is clear competitive pressure: no institution wants to fall behind in a technology that may redefine how financial services operate.

That balance between ambition and caution defines the current moment rather well. Many institutions have incorporated AI into their strategic narrative, yet only a limited number have achieved true industrial-scale…