Latest Tech News

Stay updated with the latest in technology, AI, cybersecurity, and more

Filtered by: del Clear Filter

How DeepSeek used distillation to train its artificial intelligence model, and what it means for companies such as OpenAI

Chinese artificial intelligence lab DeepSeek roiled markets in January, setting off a massive tech and semiconductor selloff after unveiling AI models that it said were cheaper and more efficient than American ones. But the underlying fears and breakthroughs that sparked the selling go much deeper than one AI startup. Silicon Valley is now reckoning with a technique in AI development called distillation, one that could upend the AI leaderboard. Distillation is a process of extracting knowledge

OpenAI bans Chinese accounts using ChatGPT to edit code for social media surveillance

OpenAI has banned the accounts of a group of Chinese users who had attempted to use ChatGPT to debug and edit code for an AI social media surveillance tool, the company said Friday . The campaign, which OpenAI calls Peer Review, saw the group prompt ChatGPT to generate sales pitches for a program those documents suggest was designed to monitor anti-Chinese sentiment on X, Facebook, YouTube, Instagram and other platforms. The operation appears to have been particularly interested in spotting call

Microsoft's Magma AI Can Manipulate and Control Robots

Microsoft just introduced Magma, a new artificial intelligence model designed to help robots see, understand and act more intelligently. Unlike traditional AI models, Magma processes different types of data all at once -- an effort Microsoft is calling a big leap toward "agentic AI," or systems that can plan and execute tasks on a user's behalf. The model, which uses a combination of vision and language processing, is trained on videos, images, robotics data and interface interactions so as to

Report: OpenAI plans to shift compute needs from Microsoft to SoftBank

In Brief OpenAI is forecasting a major shift in the next five years around who it gets most of its computing power from, The Information reported on Friday. By 2030, OpenAI expects to get three-quarters of its data center capacity from Stargate, a project that’s expected to be heavily financed by SoftBank, one of OpenAI’s newest financial backers. That represents a major shift away from Microsoft, OpenAI’s biggest shareholder, who fulfills most of the startup’s power needs today. The change w

Research shows AI will try to cheat if it realizes it is about to lose

Serving tech enthusiasts for over 25 years.TechSpot means tech analysis and advice you can trust Surprise! A recent study showed that some of the newest AI reasoning models aren't above cheating to achieve a goal. Computer scientists found that AI systems can now manipulate chess AIs to gain an unfair advantage. Some models did this without human interaction or prompting, raising concerns about the future integrity of AI-driven systems beyond the chessboard. A Palisade Research study found tha

Beijing embraces DeepSeek to lead AI adoption as it looks for new growth drivers

HONG KONG, CHINA - JANUARY 28: In this photo illustration, the DeepSeek apps is seen on a phone in front of a flag of China on January 28, 2025 in Hong Kong, China. DeepSeek's sudden splash in the large language model space has given China a powerful tool to catalyze artificial-intelligence adoption in the country and boost economic growth. While Goldman Sachs pegs a 20-basis-point to 30-basis-point boost to China's GDP over the long term — by 2030 — it expects the country's economy to start r

Posh (YC W22) Is Hiring

About POSH POSH is revolutionizing the battery energy storage industry by providing innovative, intelligent, and sustainable solutions for commercial and industrial (C&I) applications. We help businesses optimize power management and reduce carbon footprints. We are seeking an Energy Analysis & Modeling Engineer to support our mission by developing, analyzing, and optimizing energy storage models that improve system performance, efficiency, and cost-effectiveness. Responsibilities Develop, v

More than 376,000 Tesla Model Y, Model 3s have faulty steering

Some Tesla owners have yet another thing to worry about. As sales crash in Europe and protests gather outside Tesla showrooms in the US as a result of the CEO's political engagement, it now emerges that more than 376,000 Model Y crossovers and Model 3 sedans are at risk for power steering failure. So far, it has resulted in more than 3,000 warranty claims and caused 570 crashes, according to Tesla and the National Highway Traffic Safety Administration. Federal investigators have known about the

DeepSeek goes beyond “open weights” AI with plans for source code release

Last month, DeepSeek turned the AI world on its head with the release of a new, competitive simulated reasoning model that was free to download and use under an MIT license. Now, the company is preparing to make the underlying code behind that model more accessible, promising to release five open source repos starting next week. In a social media post late Thursday, DeepSeek said the daily releases it is planning for its "Open Source Week" would provide visibility into "these humble building bl

Together AI’s $305M bet: Reasoning models like DeepSeek-R1 are increasing, not decreasing, GPU demand

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More When DeepSeek-R1 first emerged, the prevailing fear that shook the industry was that advanced reasoning could be achieved with less infrastructure. As it turns out, that’s not necessarily the case. At least, according to Together AI, the rise of DeepSeek and open-source reasoning has had the exact opposite effect: Instead of reducing the need for infrastructure, it is

How test-time scaling unlocks hidden reasoning abilities in small language models (and allows them to outperform LLMs)

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Very small language models (SLMs) can outperform leading large language models (LLMs) in reasoning tasks, according to a new study by Shanghai AI Laboratory. The authors show that with the right tools and test-time scaling techniques, an SLM with 1 billion parameters can outperform a 405B LLM on complicated math benchmarks. The ability to deploy SLMs in complex reasoni

OpenInfer raises $8M for AI inference at the edge

OpenInfer has raised $8 million in funding to redefine AI inference for edge applications. It’s the brain child of Behnam Bastani and Reza Nourai, who spent nearly a decade of building and scaling AI systems together at Meta’s Reality Labs and Roblox. Through their work at the forefront of AI and system design, Bastani and Nourai witnessed firsthand how deep system architecture enables continuous, large-scale AI inference. However, today’s AI inference remains locked behind cloud APIs and host

OpenAI’s ChatGPT explodes to 400M weekly users, with GPT-5 on the way

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More OpenAI’s ChatGPT has surpassed 400 million weekly active users, a milestone that underscores the company’s growing reach across both consumer and enterprise markets, according to an X post from chief operating officer Brad Lightcap on Thursday. The rapid expansion comes as OpenAI faces intensifying competition from rivals such as Elon Musk’s xAI and China’s DeepSeek, b

OpenAI’s GPT-4.5 May Arrive Next Week, but GPT-5 Is Just Around the Corner

OpenAI may be preparing to slap a new coat of paint on ChatGPT with an updated AI model, GPT-4.5, as early as next week. If that’s not enough to get users excited, the Sam Altman-led company is on the path toward its ultimate model while trying to hint that this next step will finally achieve “AGI.” Spoiler alert: it won’t. Based on anonymous sources, the Verge’s Tom Warren first reported that OpenAI’s next model could hit the scene sometime this month. Microsoft reportedly plans to host the co

Microsoft’s Satya Nadella Pumps the Breaks on AI Hype

The world has yet to turn any of today’s AI hype and spending into a meaningful lift in the actual economy, says Microsoft CEO Satya Nadella. The comments were made during a podcast interview in which he put a measuring stick on AI’s true impact: Success will be measured through global economic growth rather than arbitrary benchmarks of how well AI programs can complete challenges like obscure math puzzles. Those are interesting in isolation but do not have practical utility. The comments seem

Microsoft’s new AI agent can control software and robots

On Wednesday, Microsoft Research introduced Magma, an integrated AI foundation model that combines visual and language processing to control software interfaces and robotic systems. If the results hold up outside of Microsoft's internal testing, it could mark a meaningful step forward for an all-purpose multimodal AI that can operate interactively in both real and digital spaces. Microsoft claims that Magma is the first AI model that not only processes multimodal data (like text, images, and vi

Apple launches $599 iPhone 16e: Best features, specs, and everything else to know

Apple It was a little more than two years ago when Apple refreshed its entry-level, low-cost iPhone model, ideal for people who didn't need all of the bells and whistles but still wanted to be part of the Apple ecosystem -- the iPhone SE. After a two-year hiatus, the next big refresh has arrived. Also: What does the 'e' in iPhone 16e stand for? On Wednesday, Apple launched the iPhone 16e via an online press release rather than the usual product event. The iPhone 16e received several massive u

Topics: 16e apple iphone model se

I’m Not Convinced Ethical Generative AI Currently Exists

Are there generative AI tools I can use that are perhaps slightly more ethical than others? —Better Choices No, I don't think any one generative AI tool from the major players is more ethical than any other. Here’s why. For me, the ethics of generative AI use can be broken down to issues with how the models are developed—specifically, how the data used to train them was accessed—as well as ongoing concerns about their environmental impact. In order to power a chatbot or image generator, an ob

Breaking down Grok 3: The AI model that could redefine the industry

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Less than two years since its launch, xAI has shipped what could arguably be the most advanced AI model to date. Grok 3 matches or beats the most advanced models on all key benchmarks as well as the user-evaluated Chatbot Arena, and its training has not even been completed yet. We still don’t have a lot of details about Grok 3, as the team has not yet released a paper

Topics: ai grok model models xai

The Forecasting Company (YC S24) Is Hiring

Your role We are building the forecasting foundation model to rule them all. All enterprise companies run forecasting to plan their operations: staffing, supply-chain management, finances… We provide the data, models and platform to easily build the most accurate forecasts. This significantly reduces waste and increases cash flow for our customers. The forecasting model is at the heart of our technology. As the founding MLE, you will build, train and deploy large foundation model architectures

Researchers Find Elon Musk's New Grok AI Is Extremely Vulnerable to Hacking

Researchers at the AI security company Adversa AI have found that Grok 3, the latest model released by Elon Musk's startup xAI this week, is a cybersecurity disaster waiting to happen. The team found that the model is extremely vulnerable to "simple jailbreaks," which could be used by bad actors to "reveal how to seduce kids, dispose of bodies, extract DMT, and, of course, build a bomb," according to Adversa CEO and cofounder Alex Polyakov. And it only gets worse from there. "It’s not just ja

Breaking down Grok-3: The AI model that could redefine the industry

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Less than two years since its launch, xAI has shipped what could arguably be the most advanced AI model to date. Grok-3 matches or beats the most advanced models on all key benchmarks as well as the user-evaluated Chatbot Arena, and its training has not even been completed yet. We still don’t have a lot of details about Grok-3, as the team has not yet released a paper

Topics: ai grok model models xai

Mistral's new AI model specializes in Arabic and related languages

chrispecoraro/Getty Images Paris-based AI startup Mistral is focusing on providing large language models (LLMs) that understand regional-specific languages and are tailored to grasp the cultural nuances sometimes overlooked in larger, more general-purpose models trained to be versed in multiple languages. Mistral has released its first "specialized" regional language-focused model, Saba. According to Mistral, the 24-billion-parameter model has been trained on "meticulously curated datasets" fr

I built a large language model "from scratch"

A developer's journey through building an LLM from scratch, sharing key insights about tokenization, training, and the learning process of mastering AI fundamentals. Building a large language model from scratch# I’m a machine learning / A.I. hobbyist. The technologies fascinate me, and I can’t seem to learn enough about them. Sebastian Raschka’s book, Build a Large Language Model (From Scratch) caught my eye. I don’t recall how I stumbled on it, but I found it when it was still in early access

Here are the best iPad deals right now

While the best iPad deals usually land during major sale events like Black Friday and Amazon Prime Day, many great iPad deals are attainable outside those times. The day-to-day discounts may come and go like changing winds, but there’s often something to be saved, particularly on the more affordable iPads. The most recent iPad Pro and iPad Air are also regularly discounted, and you can even save a bit on Apple’s new iPad Mini. It’s difficult to know where exactly you can find the most notable i