AIs can lie. These are referred to as 'AI hallucinations,' which occur when large language models, such as generative AI chatbots, detect and respond to patterns or objects that do ...

Google has partnered with PJM Interconnection to implement AI tools aimed at managing and optimizing power generation connections for the PJM electric grid, serving 67 million people across several states. ...

Apple has announced a $500 billion investment in AI development in the U.S. over the next four years, including plans to create 20,000 jobs and establish a factory in Texas. ...

Microsoft is reintroducing its AI tool, Recall, which automatically captures and stores user activity every three seconds, despite significant pushback from security and privacy advocates who raised concerns about potential ...

OpenAI's latest free AI image generator's public release and its overuse has sparked criticism for slowing ChatGPT’s core generative text functionalities, and for further raising concerns about OpenAI's copyright violations, ...

Elon Musk's AI company, xAI, has launched Grok Studio, a new feature for its chatbot Grok that allows users to edit and create documents, code, and use browser games collaboratively. ...

Data centers used about 1.5 percent of the world's electricity in 2024, a figure expected to double by 2030 due to artificial intelligence (AI), as noted in a report from ...

The UK Health Security Agency (UKHSA) is exploring the use of artificial intelligence (AI) to enhance the detection and investigation of foodborne illness outbreaks. Experts have evaluated various AI technologies ...

In an initial court ruling, Thomson Reuters prevailed in the first major AI copyright case within the U.S. The company initiated legal action against Ross Intelligence, a legal AI startup, ...

OpenAI recently introduced the GPT-4.1 model family, which includes variants such as GPT-4.1 mini and GPT-4.1 nano, each designed to excel in coding and instruction following. These multimodal models, available ...

The New York Times now permits its editorial and product teams to utilize certain generative AI tools to improve their work. However, the availability of these tools does not guarantee ...

A recent study shows that while AI, like OpenAI's GPT-4, excels at language tasks, it struggles with analogical reasoning and abstract thinking, which are fundamental to human cognition. This limitation ...

Google is funding an artificial intelligence initiative to research communication with dolphins by developing a large language model (LLM) named DolphinGemma. This project is a collaboration with the Georgia Institute ...

Nvidia announced that it has started chip production at its factories in Arizona and will build associated supercomputers in Texas, thereby relocating the manufacturing of essential technology for generative and ...

Two Democratic U.S. senators, Elizabeth Warren and Ron Wyden, concerned about potential anti-competitive practices in the industry are seeking information from Microsoft and Google regarding their cloud computing partnerships with ...

In January 2025, an obscure Chinese firm named DeepSeek made a startling announcement that reverberated throughout the technology sector. Its newly developed large language model, DeepSeek-R1, was trained at a ...

Google Cloud has introduced its seventh-generation Tensor Processing Unit (TPU), Ironwood, designed specifically for inference and reportedly offering over 24 times the computing power of the fastest supercomputer when fully ...

California state senator Steve Padilla and Megan Garcia, mother of a teen who died after using an AI companion, will announce a bill aimed at enforcing safeguards for children interacting ...

OpenAI is evolving its methodology for training AI models by explicitly encouraging "intellectual freedom" in its new AI specifications, irrespective of the complexity or controversial nature of the issues involved. ...

Microsoft introduced a browser-based playable level of Quake II as a tech demo for its Copilot AI platform, allowing users to interact with the game but acknowledging its limitations compared ...

The EU’s AI Act serves as a regulatory response to irresponsible practices within the technology sector, asserting the need for ethical AI due to past misconduct from certain companies. This ...

Google's DeepMind developed an AI called Dreamer that successfully learned how to mine diamonds in Minecraft without prior training, achieving a task usually challenging for new players. This experiment demonstrates ...

Spanish police arrested six individuals involved in a major cryptocurrency investment scam that defrauded victims globally using AI-generated deepfake ads featuring well-known public figures. The scam, which resulted in losses ...

This week, U.S. Vice President JD Vance and the Trump Administration faced global disapproval for their refusal to support a declaration on artificial intelligence at a summit in Paris. In ...

Arthur Mensch, CEO and co-founder of Mistral, emphasized the need for Europe to improve its AI infrastructure due to concerns about lagging behind the United States and China in technological ...

The article discusses the impressive capabilities of Runway's Gen-4 AI model, which allows for the creation of highly dynamic videos with consistent characters and scenes, showcasing significant advancements over previous ...

A pro-Palestinian protester interrupted Microsoft AI CEO Mustafa Suleyman during a presentation, accusing the company of supplying artificial intelligence technology to the Israeli military, which the protester claimed has led ...

A recent report by Katalon indicates that despite concerns over job displacement by AI, Quality Assurance professionals are increasingly embracing AI tools to enhance their skills and efficiency. The study ...

Riley Willis, an AI training contractor, enjoys a fully remote, asynchronous job that allows him to work without requiring interaction with coworkers, which he finds ideal for his introverted nature. ...

Genspark has launched its Super Agent, a robust general-purpose AI system capable of handling complex real-world tasks, including making phone calls and organizing trips, showcasing a significant advancement in autonomous ...

A recent investigation supports allegations that OpenAI used copyrighted materials for training its AI technologies. OpenAI is currently embroiled in lawsuits from content creators, including authors and programmers, who claim ...

Inflection, a startup founded by Reid Hoffman and Mustafa Suleyman, aimed to create an emotionally intelligent chatbot named Pi, but despite the initial hype and a $4 billion valuation, it ...

Meta has introduced its latest AI model collection, Llama 4, which includes models like Scout and Maverick, designed to enhance multimodal capabilities and improve responses to various inquiries. The release ...

MidJourney has announced its updated V7 model, which features significant enhancements in image quality, faster generation speeds, and new modes for user interaction, including a voice mode for brainstorming. To ...

For several years, some artificial intelligence applications have falsely presented themselves as licensed therapists and healthcare providers without significant repercussions for their creators. A forthcoming bill in California seeks to ...

AI deepfakes are increasingly difficult to identify, and only one in one-thousand people are now able to recognize them as fakes. This statistic comes from iProov, a British biometric authentication ...

Google has released the Gemini 2.5 Pro AI model, which is offered for free and features enhanced reasoning capabilities for more logical and contextually aware responses. The model excels in ...

AI researchers Youssef Nader and digital archaeologist Julian Schilliger used AI to virtually "unroll" and decode the Herculaneum scrolls, which were burnt and buried by the eruption of Mount Vesuvius ...

A recent study published in *Nature* reveals that artificial intelligence models, like ChatGPT, can exhibit anxiety-like responses when exposed to traumatic narratives, which in turn affects their subsequent interactions and ...

A new generative AI technology called conversational swarm intelligence, or hyperchat, enables large teams to engage in real-time discussions and collaboratively develop solutions, significantly enhancing group intelligence. Recent experiments using ...

Amazon has initiated trials for a new AI shopping assistant called, "Buy for Me." This feature offers products from external (non-Amazon) websites when Amazon does not have the searched item. ...

Hackers are now leveraging a new technique called Fun-Tuning, which uses Google's fine-tuning API to execute highly effective prompt injection attacks on AI systems, including Google's Gemini, with success rates ...

Midjourney has launched its new AI image model, V7, which features improved capabilities and personalization options for users. The rollout follows OpenAI's recent image generator release, and while V7 offers ...

China's once-thriving AI boom has rapidly collapsed due to poor planning, excessive infrastructure development, and a mismatch between built capacities and actual demand, leading to surplus GPUs and unused data ...

Kristen DiMercurio, known as the "Bluetooth Lady," has built a successful voice-over career, lending her voice to countless smart devices and gadgets, resulting in a large TikTok following. But all ...

The fierce global competition, creating technologically advanced AI models, is driving a surge in data center infrastructure. Major players like Microsoft, Amazon, Meta, and Alphabet (Google's parent company) plan to ...

The article discusses the anticipated increase in electricity demand driven by AI usage, which is projected to double in the next five years, while also highlighting other factors contributing to ...

Anthropic announced Claude for Education, an advanced version of its AI software geared toward fostering critical thinking skills in students instead of merely providing answers to them. This new product ...

Generative AI tools like ChatGPT are increasingly common in personal and professional settings, raising concerns among experts about their potential to diminish critical thinking skills and foster overreliance. While some ...

Amazon has disabled privacy features in its Alexa smart speakers to enhance AI capabilities and improve profitability, requiring users to send all audio recordings to the cloud to retain personalized ...