Notizie — Pagina99

OpenAI and Anthropic announced new measures aimed at identifying and protecting underage users of their AI chat services. OpenAI will incorporate four new principles for users under 18 into its Model Spec, prioritize teen safety, and begin testing an age‑prediction system that will automatically apply safeguards when a user is flagged as likely underage. Anthropic, which already blocks users under 18 from accessing Claude, is developing a detection system that looks for subtle conversational cues indicating a minor. Both companies cite ongoing pressure to improve safety after recent legal challenges and public scrutiny.Leggi di più

Large language models (LLMs) are driving a surge in scientific manuscript submissions, especially among researchers with limited English proficiency. Papers generated with LLM assistance display higher linguistic complexity, yet they are less likely to reach peer‑reviewed publication. The AI‑generated work also cites a broader array of sources, including more books and recent papers, suggesting a diversification of references. Researchers caution that many manuscripts may still undergo extensive human editing, and that recent drafts could be under‑represented in publication metrics, potentially skewing the observed effects.Leggi di più

The input provided does not contain any source material to aggregate into a news story. Without explicit facts, quotes, numbers, or details, it is not possible to create a factual report. Consequently, this summary acknowledges the absence of source content and explains that no substantive story can be generated based on the current input.Leggi di più

Haotian, a Chinese‑language artificial‑intelligence app that offers near‑perfect face‑swapping for live video calls, has become a key tool for fraudsters in Southeast Asia. The service, promoted mainly through a Telegram channel, enables users to adjust facial features and even clone voices in real time. Researchers found that Haotian’s customers are increasingly scammers who use the technology for "pig‑butchering" schemes, cryptocurrency fraud, and other illegal activities. Payments to the company, largely in USDT, total about $3.9 million and include funds linked to known fraud cases. Haotian’s ties to the now‑sanctioned escrow platform Huione Guarantee further embed it in the region’s gray‑market infrastructure.Leggi di più

OpenAI has introduced an app platform within ChatGPT, allowing users to access popular digital services such as Adobe Photoshop, Spotify, Canva, and Zillow directly through the chatbot. The storefront categorizes apps into Featured, Lifestyle, and Productivity sections, and users connect apps via a simple "Connect" button that enables @‑prompt interactions. While the desktop version shows the new apps, the mobile iOS version has not yet received them. The rollout follows OpenAI's Dec. 17 announcement that developers could submit apps, and comes amid a lawsuit from Ziff Davis alleging copyright infringement.Leggi di più

Luma AI, the a16z-backed AI video and 3D model company, announced a new model called Ray3 Modify that lets creators edit existing footage by providing character reference images and start‑and‑end frames. The model preserves the original motion, timing, eye line and emotional delivery while allowing transformations such as costume changes, location swaps or full reshoots with AI. Ray3 Modify is available through Luma’s Dream Machine platform and follows a recent $900 million funding round led by the Public Investment Fund‑owned AI firm Humain.Leggi di più

Director Adam Bhala Lough set out to make a documentary about OpenAI CEO Sam Altman but received no response to his interview requests. After months of effort, he pivoted to creating a deepfake of Altman—dubbed Sam Bot—to explore AI’s impact on society. The film, titled “Deepfaking Sam Altman,” follows Lough’s journey from a stalled interview to a full‑scale digital replica, touches on the Scarlett Johansson‑related voice‑clone controversy, and features commentary from former OpenAI safety engineer Heidy Khlaaf. Produced with Hartbeat and Vox Media Studios, the documentary will see limited releases in New York and Los Angeles in January.Leggi di più

Google has made its Gemini 3 Flash model the default AI behind both the Gemini app and the AI Mode in Google Search. The new model is positioned as faster and more capable than its predecessors, offering users curated answers while still providing source links. Early testing shows Gemini 3 Flash can assemble budget‑friendly product bundles, troubleshoot technical issues, and compare photo‑storage options with speed comparable to traditional search. While classic Google Search continues to deliver breadth and raw source material, the AI‑driven mode emphasizes clarity, context, and time‑saving, suggesting a complementary role for both approaches.Leggi di più

OpenAI has introduced an official app store for ChatGPT, gathering a range of third‑party applications that extend the AI’s capabilities. Users can browse categories, search for specific tools, and add apps such as Apple Music, Photoshop, Canva, Gmail and AllTrails directly within the ChatGPT interface. Each listing includes a privacy policy and requires OpenAI approval before appearing to users. OpenAI is also inviting developers to submit new apps that bring fresh context and actions to conversations, promising a growing ecosystem of AI‑enhanced services.Leggi di più

Google has woven its Gemini AI directly into Maps, replacing the old Assistant button with a new Gemini button. Users can ask the app for coffee shops, scenic lunch spots, traffic updates and alternative routes, all in conversational language. The AI draws on Maps’ business listings, reviews, images and real‑time traffic data to provide context‑aware suggestions that adapt as you move. While the system occasionally adds a few seconds of latency, it delivers more nuanced, multi‑criteria recommendations than traditional map searches, making navigation feel more natural and proactive.Leggi di più

OpenAI introduced GPT-Image-1.5, a new AI image model that is four times faster than its predecessor and offers improved instruction following, editing tools, and legible text generation. The model is now accessible to all ChatGPT users worldwide through a dedicated sidebar space and is also available via API. The rollout comes as OpenAI competes with other advanced image generators, including Google's Nano Banana Pro, and follows recent partnership announcements with Disney.Leggi di più

Experts warn that the European Union could use its control of Dutch chip‑making equipment and stricter data‑privacy enforcement as a strategic lever against the United States. ASML’s monopoly on advanced lithography machines, vital for companies like Nvidia, gives Brussels the ability to restrict exports and influence the U.S. AI economy. Simultaneously, the EU’s push to enforce GDPR against firms such as Meta and Google could impede the data flows that power AI models. Analysts suggest that combined pressure on technology supply and data usage could destabilize the AI investment surge and create political challenges for President Trump.Leggi di più

OpenAI has introduced GPT Image 1.5, an AI image‑synthesis model that runs within the same neural network that processes language prompts. The new model is reported to generate images up to four times faster and at roughly 20 percent lower cost than its predecessor. Integrated into ChatGPT, GPT Image 1.5 enables users to edit photos with simple text commands, from adding objects to changing clothing, while preserving facial likenesses. The rollout marks a shift toward more seamless, conversational image editing without needing specialized graphics skills.Leggi di più

Google announced that its Opal AI‑driven app‑building tool is now embedded directly within the Gemini web experience. The move lets users design custom, task‑specific mini‑apps—called Gems—by describing desired functionality in natural language. Opal translates those prompts into step‑by‑step workflows, offering a visual editor that requires no code. Existing Gems include learning coaches, brainstorming assistants, career guides, coding partners and editors, while users can also create and reuse their own. The integration expands Gemini’s utility as a generative AI hub and highlights Google’s push toward consumer‑friendly AI development tools.Leggi di più

Amazon is reportedly in preliminary discussions to invest as much as $10 billion in OpenAI, a deal that would see the AI lab using Amazon's AI chips. The potential agreement could value OpenAI at more than $500 billion. The move follows Amazon's earlier $8 billion investment in Anthropic and its recent launch of the Trainium chip series. OpenAI's shift to a for‑profit model has opened the door to new partnerships beyond its long‑time backer Microsoft. The proposed investment would add to a series of circular deals linking AI startups with hardware and cloud providers.Leggi di più

Google unveiled Gemini 3 Flash, a faster and cheaper AI model built on the recent Gemini 3 architecture. The company is making Flash the default model in its Gemini app and AI‑enabled search, while still offering the Pro version for more demanding tasks. Gemini 3 Flash delivers notable performance gains on benchmark tests, supports multimodal inputs such as video, sketches, and audio, and is available through Vertex AI, Gemini Enterprise, and an API preview. Early adopters like JetBrains and Figma are already integrating the model, and Google highlights its suitability for bulk, work‑horse workloads.Leggi di più

OpenAI has named former British Chancellor George Osborne as Managing Director and Head of OpenAI for Countries, a role focused on expanding the company’s Stargate data‑center network beyond the United States. Osborne will guide government partnerships, help nations adopt OpenAI tools, and oversee the first round of collaborations with ten countries. His appointment underscores OpenAI’s strategy to deepen ties with governments and accelerate the rollout of sovereign AI infrastructure worldwide.Leggi di più

Google has launched Gemini 3 Flash, a new AI model integrated into its AI Mode that promises search‑engine speed without sacrificing the advanced reasoning capabilities of the Gemini 3 family. The model is available worldwide for free, and Google says it can handle complex queries with greater precision while maintaining the quick response times users expect from traditional search. Alongside Gemini 3 Flash, the company is expanding access to Gemini 3 Pro and the Nano Banana Pro image‑generation tool, though higher usage limits require a Google AI Pro or Ultra subscription.Leggi di più

Amazon is in advanced discussions with OpenAI about a $10 billion investment that would also see the AI startup adopt Amazon's Trainium chips and increase its use of Amazon Web Services. The proposed deal could lift OpenAI's valuation past $500 billion while deepening the company's reliance on Amazon's cloud and hardware infrastructure. At the same time, OpenAI continues to expand relationships with other technology partners, including Microsoft, NVIDIA, Oracle and AMD, prompting investors to scrutinize the circular nature of its financing and compute arrangements.Leggi di più

Google has launched Gemini 3 Flash, a new generative AI model that promises faster performance and higher accuracy than its predecessors. Available now through the Gemini app, search, Gemini API, Vertex AI, AI Studio, and Antigravity, the model delivers notable gains on academic, reasoning, and coding benchmarks while offering lower token costs. Gemini 3 Flash narrows the gap with the larger Gemini 3 Pro, delivering comparable results on many tests at a fraction of the price, positioning it as a versatile option for developers and users seeking efficient, high‑quality AI outputs.Leggi di più