Latest AI & Business News
Stay updated with the latest insights in AI and business, delivered directly to you.
-
Google releases responsible AI report while removing its anti-weapons pledge
The company’s annual reflection on safe AI development comes amid shifting guidance around military AI.
-
‘The biggest heist in American history’: DC is just waking up to Elon Musk’s takeover
Demonstrators hold signs during a protest against Elon Musk outside of the US Department of Treasury building in Washington, DC, on Tuesday, February 4th, 2025. The long block outside of the US Treasury Department on Tuesday was lined with protesters. They flooded across the street with signs bearing slogans: âNobody elected Muskâ; “This is illegal”;…
-
Herman Miller made a convertible standing desk that actually looks great
The smallest version of Herman Miller’s new Spout sit-to-stand tables starts at $2,300. | Image: Herman Miller Herman Miller has announced a new height-adjustable desk that doesn’t look like one. Instead of relying on just two columns to raise and lower its work surface, the Spout Sit-to-Stand Table uses four cylindrical legs, each containing their…
-
AI to shake up Salesforce workforce with possible shift to sales over IT
Salesforce apparently wants to cut 1,000 jobs to hire new staff elsewhere as part of an effort to bring its new AI products to customers, according to financial news service Bloomberg, citing internal sources. Employees who are being laid off can apply for other positions in the software company, Bloomberg reported. It is not known…
-
iOS App Store apps with screenshot-reading malware found for the first time
Apps distributed through both Apple and Google’s app stores are hiding malicious screenshot-reading code that’s being used to steal cryptocurrency, the cybersecurity software firm Kaspersky reported today. It’s the “first known case” of apps infected with malware that uses OCR tech to extract text from images making it into Apple’s App Store, according to a…
-
Meet Satori: A New AI Framework for Advancing LLM Reasoning through Deep Thinking without a Strong Teacher Model
Large Language Models (LLMs) have demonstrated notable reasoning capabilities in mathematical problem-solving, logical inference, and programming. However, their effectiveness is often contingent on two approaches: supervised fine-tuning (SFT) with human-annotated reasoning chains and inference-time search strategies guided by external verifiers. While supervised fine-tuning offers structured reasoning, it requires significant annotation effort and is constrained by…
-
‘Dengue Boy’ Is the Weird, Fleshy Novel You Need Right Now
Dengue Boy, a book about a humanoid mosquito taking his revenge in the dying years of planet Earth is unsettling and essential.
-
Longer-lasting laptops: the modular hardware you can upgrade and repair yourself
When it comes to improving the sustainability of consumer electronics, there’s a growing movement to design devices with a focus on upgradability and repairability that can extend their longevity, instead of just making them easier to recycle after a few years of use. At one time, nearly every laptop maker was caught up in a…
-
Nintendo says it’s ‘taking risks’ to meet Switch 2 demand
Nintendo is gearing up for the Switch 2’s launch sometime in 2025, and president Shuntaro Furukawa says that the company is “taking risks and proceeding with production so that we can meet as much demand as possible,” according to a machine translation of an earnings Q&A. The company had some trouble keeping the original Switch…
-
Not every AI prompt deserves multiple seconds of thinking: how Meta is teaching models to prioritize
Let models explore different solutions and they will find optimal solutions to properly allocate inference budget to AI reasoning problems.Read More