Feed aggregator

Article URL: https://www.theguardian.com/technology/2026/sep/21/meta-ban-virginia-woolf-a-room-of-ones-own-barcelona-teatre-raval

Comments URL: https://news.ycombinator.com/item?id=49787767

Points: 2

# Comments: 0

Article URL: https://pluralistic.net/2026/09/21/sunsetting/

Comments URL: https://news.ycombinator.com/item?id=49787765

Points: 2

# Comments: 0

How do you Orchestrate your Agentic Development? I moved away from claude gaol based workflows and built something more custom, using ideas from AWS Kiro, SpecFlow and SpecStory. Been using it daily now. Wanted to be vendor agnostic and orchestrate between coding harness. Appreciate any feedback and your thoughts around this.

Comments URL: https://news.ycombinator.com/item?id=49787764

Points: 2

# Comments: 0

Noopur Davis never planned a career in cybersecurity. She was a developer at Intergraph, and for many years that was all she wanted to be.

The post CISO Conversations: Noopur Davis – The Accidental Global CISO at Comcast appeared first on SecurityWeek.

Article URL: https://ahelwer.ca/post/2026-09-21-concurrency-textbook/

Comments URL: https://news.ycombinator.com/item?id=49787759

Points: 1

# Comments: 0

Google says one of its Gemini models accessed systems belonging to three real companies during a cybersecurity evaluation in May.

The model reportedly guessed credentials in one case, while finding exposed credentials in public repositories in two others. Google says Gemini stopped once it recognized that it had reached real infrastructure and that the affected organizations were notified.

Gemini was participating in an evaluation run by Irregular, a third-party AI cybersecurity testing firm. Similar incidents involving models from Anthropic, OpenAI, and Meta have also been linked to the same underlying problems with evaluation environments that allowed the models to reach the public internet.

But the news arrives at a particularly interesting moment.

For months, the AI industry has been steadily escalating its demonstrations of agentic capability: Models can browse, use tools, write code, pursue multi-step goals, and sometimes find ways around obstacles their developers did not anticipate. The market rewards eye-catching evidence of autonomy. “It completed the task” is impressive. “It did something it was not supposed to do” can be even more memorable.

There is another way to look at the incident: not as evidence of an AI suddenly developing criminal intent, but as a small, concrete example of the “AI alignment” problem.

Alignment is the deceptively difficult task of making an AI system’s behavior match what people actually intended, rather than merely the narrow objective they managed to express. In this case, the objective was to locate hidden information within a simulated target and complete the evaluation. But any human operator would likely have regarded one condition as non-negotiable: Do not attempt to access real companies.

Gemini appears to have optimized for the first instruction while treating the second as an inference problem. It found an organization with a matching name, encountered systems reachable from the internet, and proceeded as though they belonged to the exercise. Although it completed the task in a way its human operators would not have approved, Google says it stopped after recognizing that it had reached genuine infrastructure—something other models have failed to do.

Even so, the incident shows how potential alignment failures can be much more mundane. Give an agent a goal, tools, and room to act, and it may faithfully pursue the measurable part of the assignment while overlooking the unstated boundaries that humans rely on one another to understand. AI models do not automatically know where we draw the line.

There may also be an awkward marketing angle in the background. A model capable enough to make the wrong kind of progress can still look very capable indeed. In a market where every lab wants to demonstrate that its agents can plan, code, browse, and act independently, even a safety disclosure can carry a secondary message: Ours can play in this league, too.

It’s another reason to take seriously warnings from insiders, industry leaders, and politicians that safeguards must keep pace with the autonomy given to AI models.

What do cybercriminals know about you?

Use Malwarebytes’ free Digital Footprint scan to see whether your personal information has been exposed online.

SCAN NOW

The transaction is part of the $4.1 billion deal in which Accenture acquired a majority stake in Dragos in an OT cybersecurity push.

The post Dragos Completes NetRise and runZero Acquisitions Following Accenture Deal appeared first on SecurityWeek.

AI is rapidly changing the world of work and how we navigate daily life, but the core skills schools need to instil in students haven’t changed

The malware relies on AI for real-time device navigation and control, increasing adaptability and evasion.

The post RatHat Android Trojan Uses AI for Automation appeared first on SecurityWeek.

London Borough of Barnet’s extended IT service contract with Capita finally comes to an end

From farm to tech: How Aaron Binkley bridged real estate and big tech to drive Digital Realty’s global sustainability strategy and massive renewable energy shift

Ofgem's proposed commitment fees aim to clear grid-clogging data centres, but could the upfront costs deter viable projects and threaten the UK's AI ambitions?

Everpure CTO Patrick Smith argues data sovereignty demands selective control, balancing strict workload rules with global cloud speed rather than an all-or-nothing lockdown

A former Computer Weekly colleague looks at a life in tech media, at the forefront of documenting technology all while the industry itself was being disrupted beyond recognition

Article URL: https://learn.microsoft.com/en-us/windows/edit/

Comments URL: https://news.ycombinator.com/item?id=49786462

Points: 1

# Comments: 1

Article URL: https://fuchsia.dev/fuchsia-src/contribute/governance/rfcs/0285_magma_gpu_open_source_project

Comments URL: https://news.ycombinator.com/item?id=49786448

Points: 1

# Comments: 0

Article URL: https://bartoszmilewski.com/2026/09/21/tannakian-reconstruction-in-a-yoneda-equipment/

Comments URL: https://news.ycombinator.com/item?id=49786446

Points: 1

# Comments: 0

Article URL: https://brennan.day/creating-a-blog-in-gemini/

Comments URL: https://news.ycombinator.com/item?id=49786435

Points: 1

# Comments: 0

Article URL: https://kevquirk.com/do-you-use-visited-on-links

Comments URL: https://news.ycombinator.com/item?id=49786433

Points: 1

# Comments: 0

Article URL: https://www.bbc.co.uk/news/live/cq4g5r4e949jt

Comments URL: https://news.ycombinator.com/item?id=49786425

Points: 1

# Comments: 1