Tech News
← Home  ·  All topics

Ai

5382 GoKawiil briefs on this topic · RSS for ai

DEEPX touts DX-M1 AI chip's cooler running in CNBC butter demo

South Korean startup DEEPX showcased its DX-M1 AI processor in a CNBC segment using a demonstration where blocks of butter placed atop rival chips melted while butter on the DX-M1 stayed solid. CEO Kim Nok-won claimed the chip runs 25 to more than 40 degrees Celsius cooler than unnamed competitors while processing the same AI workload.

Essay Argues Engineers Still Outperform Peers Despite AI Agent Takeover of Coding

A software engineer writes that AI agents now handle most coding, debugging, documentation, and even architecture work, leaving engineers mainly coordinating multiple agents so they don't block each other. Using a proof-by-contradiction argument, the author notes that if humans added no value, skilled and average engineers would produce similar results with the same models and tools — but the gap between them persists, and top performers have become even more productive.

CNET Guide Shows How to Use ChatGPT, Gemini and Claude for Interview Prep

CNET published a guide detailing how job seekers can use AI chatbots like ChatGPT, Gemini and Claude to rehearse for interviews, including running mock interviews with voice mode, generating company-themed Zoom backgrounds, and researching expected salary ranges. The guide also notes chatbots can help prepare answers to behavioral questions and research potential employers.

Report: Apple weighing faster rollout of agentic Siri features for 2026

Bloomberg's Mark Gurman reported that Apple is considering giving Siri agentic capabilities—the ability to pursue multi-step goals without constant user prompts—as soon as next year. A separate Gurman report said Apple CEO John Ternus has been pushing software chief Craig Federighi to accelerate development timelines, including work beyond basic Siri fixes.

Trump claims he threatened to jail an unnamed AI company executive

In a CBS '60 Minutes' interview, Donald Trump said he personally warned a top AI lab executive to 'get ready to go to jail' over an unspecified issue described as 'doing something bad with AI.' He did not name the executive or detail what the alleged wrongdoing was, saying only the person had a wife and children.

Report details GPU, CPU and NPU benchmarks for running local AI models

New testing data compares how different hardware handles local large language models, measuring output in tokens per second. Nvidia's RTX 5090 generated 205 tokens per second on gpt-oss-20b, Apple's M4 chip produced about 24 tokens per second on a 7B model, and Nvidia's DGX Spark ranged from 38 tokens per second on Llama 3.1 8B down to 4.42 on its 70B version. Separate hands-on testing on a 2024 Asus Zenbook Duo showed Intel's OpenVINO toolkit offloading work to an NPU kept the laptop responsive, versus heavy CPU and memory strain when running the same model through Ollama.

Instagram launches Effect Builder to create AI-generated Stories filters

Instagram has released Effect Builder, a tool that lets users generate custom Stories filters using AI prompts, currently available to US residents over 18. Users can type a description or attach reference images, preview the effect on their own photos, and refine it through follow-up prompts, with an option for Instagram's AI to expand their request into more detailed styling instructions. Approved filters pass a safety review before becoming available to all users, with credit linking back to the creator's profile.

Report urges companies to turn AI experiments into shared organizational practice

A new commentary argues that while AI adoption inside companies is accelerating, individual employees experimenting with AI tools has not translated into broader workplace transformation. It calls on organizations to identify what works for individuals and scale those successes across teams rather than leaving gains isolated.

Enterprise leaders urged to pair AI scale-up with formal governance frameworks

A commentary piece argues that companies seeking to expand AI programs should do so by establishing clear trust, defined roles and decision-making authority rather than scaling AI deployments without structure. The author draws on experience building C-suite relationships to argue governance should underpin AI investment decisions.

Envision's 800MW Inner Mongolia AI campus built at fraction of US data center cost

SemiAnalysis reports that Chinese company Envision has completed the first 120MW phase of its Galaxy AI data center campus in Ulanqab, Inner Mongolia, in under a year, at roughly one-fifth the cost of comparable US facilities. The full campus, serving two unnamed internet companies, is planned to reach 800MW of IT capacity by 2027, pairing data halls with dedicated wind farms and battery storage rather than relying solely on the local grid.

Lubrizol uses cross-functional 'fusion teams' and cloud AI to speed chemical R&D

Lubrizol is combining scientists, data scientists, and IT staff into joint 'fusion teams' to apply AI and cloud computing to chemical research, starting with small pilot projects. The company's Rao says this approach is being used to model new formulations, including sustainable alternatives like starch and corn-based materials, and to help scale lab discoveries into manufacturing.

Meta expands AI age-estimation for Instagram teen accounts worldwide

Meta is rolling out its AI-based age estimation system globally to catch teens who lie about their age and automatically place them into Instagram's Teen Account protections. The company says US parents will start receiving in-feed notifications this month, linking to a video and an article explaining the age-assurance approach.