The Morning Download: AI’s Unspoken Risk

Good morning. Beyond the mounting concerns about a range of AI-related risks, there’s a more fundamental story in play. Frontier models are still advancing in a meaningful way, and every new generation alters the potential impact of AI on the enterprise.

This isn’t to say that those anxieties aren’t valid and don’t demand attention. The Trump administration forced Anthropic to impose new cybersecurity controls on its leading-edge model, which is now available for use after it was pulled from the market several weeks ago. From the Wall Street Journal’s coverage:

Under a deal announced Tuesday evening, Anthropic would address the workarounds that researchers at Amazon used to evade the safeguards for Fable, a public version of Anthropic’s powerful Mythos model that is capable of carrying out cyberattacks, Commerce Secretary Howard Lutnick said on X. The guardrails are critical for the company to release the model publicly.

And the Bank of International Settlements, based in Basel, Switzerland, warned Sunday that fierce competition risks driving AI investment spending to excessive levels, threatening the profitability of leading firms and a sharp reversal that could tip some economies into recession, as the WSJ reports:

“The race to capture market share may have led to overinvestment,” said Pablo Hernández de Cos, general manager of the BIS. “This could leave the sector more vulnerable if AI under delivers, possibly bringing the current investment boom to an abrupt end.”

Tech Leader Takeaway. The more elusive risk is how these evolving models will shape the operating environment for companies. The development of one generation of model to the next might not seem dramatic when it’s expressed as an improvement in an engineering benchmark, but people experience them as significant changes.

We often refer to models as commodities, but I don’t think that’s the case as long as they’re powerful enough to compel users to switch to something new. The greatest AI-related risk for companies is that six months from now, models will be materially better than they are today, stressing the organization’s ability to adapt to a constant and accelerating pace of change.

• Anthropic and the Trump administration struck a deal to restore access to Fable, the public version of its Mythos model, after a roughly 2.5-week shutdown triggered by security concerns, with Anthropic agreeing to fix the workarounds Amazon researchers used to evade Fable's safeguards, WSJ reports. Anthropic said it is also working with other tech giants on a consensus framework for gauging jailbreak severity and how developers should respond.

• Anthropic on Tuesday announced Claude Science, software designed to help scientists in molecular and cellular biology and other fields automate tasks like protein structure prediction, Bloomberg reports. The company also shared plans to pursue its own preclinical drug discovery program. The combined news sent shares of companies involved in drug discovery down sharply, Bloomberg said.

• Microsoft is planning for cuts in its Xbox gaming division as well as sales and consulting jobs cuts, impacting less than 2.5% of its global workforce, Business Insider reports. 

• Amazon Web Services said it was investing $1 billion to create a unit of forward deployed engineers to help customers develop and deploy AI. The setup plays off a concept made popular by Palantir and more recently adopted by OpenAI of placing AI experts directly with customers. AWS said its FDE unit is focused on agents and "designed so customers are self-sufficient when a deployment ends.”

• Tech workers in San Francisco tell the New York Times that a $180,000 salary is barely getting by and now with OpenAI and Anthropic preparing for IPOs the perfect apartment is about to get a whole lot more expensive.

This September 14–15, technology leaders will gather in New York City for the WSJ Technology Council Summit to explore how enterprise AI is moving from experimentation to measurable business value. Join the Technology Council and be part of the conversations shaping the future of leadership, as executives tackle AI deployment, cybersecurity, evolving technology policy, enterprise transformation and the strategies driving the next generation of business innovation.

Request an Invitation

Follow Isabelle Bousquette on LinkedIn, Instagram, X, and TikTok for more behind the scenes on her tech and AI coverage, and lately, her contributions to the WSJ Leadership Institute's new Executive Resilience series, where she's profiling America's top execs about their fitness and wellness habits.

Follow Belle Lin on LinkedIn and X for her latest reporting on enterprise technology and AI.

Steven Rosenbush is chief of the enterprise technology bureau at the WSJ Leadership Institute. He also has a column. You can follow him on LinkedIn.

Tom Loftus is the editor of The Morning Download. He suggests following Isabelle, Belle and Steve on their various social channels. But if you insist, here's his LinkedIn.