Notizie — Pagina 2

OpenAI announced Wednesday that it has created its first custom‑designed inference processor, named Jalapeño, in partnership with Broadcom. The chip, optimized for running pre‑trained AI models, promises markedly better performance‑per‑watt than existing solutions and could lower the cost of real‑time AI services. While still in testing, the move signals OpenAI’s intent to lessen reliance on Nvidia GPUs and to control more of the hardware stack that powers its products.Leggi di più

Google has integrated its standalone computer-use model into Gemini 3.5 Flash, the AI agent launched at I/O 2026. The new built‑in capability lets agents see, click, type and scroll across browsers, mobile devices and desktops without a separate model. Google pairs the feature with optional enterprise safeguards that require user confirmation for sensitive actions and automatically halt execution on suspected prompt‑injection attacks. The move aims to simplify multi‑step automation for developers while emphasizing security in regulated environments.Leggi di più

Amazon MGM announced it will no longer distribute Luca Guadagnino’s biographical drama “Artificial,” which chronicles OpenAI co‑founder Sam Altman’s 2023 ouster and reinstatement. The decision follows weeks of studio rejections, with Netflix, A24, Focus Features and Warner Bros. Clockwork all declining to pick up the film. Amazon cited a belief the project would be “better served” elsewhere, a statement that comes as the tech giant deepens its AI investments. With the film’s release plans for early 2027 and SXSW now in limbo, the future of the high‑profile biopic remains uncertain.Leggi di più

ByteDance announced Seedance 2.5 at its Volcano Engine FORCE conference in Beijing, introducing a 30‑second native 4K video generator that accepts up to 50 multimodal reference inputs. The upgrade skips four intermediate versions, positioning the model as a production‑grade tool rather than a novelty. An enterprise beta is already live, with a public launch slated for early July. The rollout follows a pause after Disney, Warner Bros., Paramount and Netflix demanded watermarking and IP safeguards for the earlier Seedance 2.0, and it arrives amid a shifting AI video market.Leggi di più

Indian customer‑engagement platform MoEngage has bought San Francisco‑based AI startup Aampe in an all‑cash transaction worth several tens of millions of dollars. The acquisition gives MoEngage a dedicated AI‑agent technology that tailors messages to individual shoppers, a move the company says will help it win business from rivals like Salesforce and Adobe. Around 20 Aampe staff will join MoEngage, boosting its headcount to roughly 820 employees.Leggi di più

Superhuman announced Tuesday that it has bought GPTZero, the Princeton‑origin AI‑detection platform that grew to more than 19 million users and $30 million in annual recurring revenue. The terms of the deal were not disclosed. The acquisition adds GPTZero’s technology to Superhuman’s existing AI‑detector, a move the company says gives users “two AI detectors are better than one.” GPTZero’s founders, Edward Tian and Alex Cui, previously raised $13.5 million from investors including Uncork Capital and Footwork.Leggi di più

MGX, the Abu Dhabi‑based investment firm built for the AI era, closed a $50 billion fund that draws money from regional sovereign wealth funds, global pension plans and other institutional investors. The fund, one of the largest ever raised solely for artificial‑intelligence investments, is already backing frontier‑model developers, chip makers and data‑center operators, while its chairman Sheikh Tahnoon bin Zayed Al Nahyan says the firm aims to surpass $100 billion in assets under management within years.Leggi di più

Irish AI‑inference startup TensorX has raised €8 million to purchase Nvidia’s latest Blackwell B300 GPUs and broaden its Europe‑focused, GDPR‑compliant service. The funding, supplied by Darius Cubed Ventures, will fund hardware upgrades at its Dublin and Helsinki data centres and support a rollout to additional sites across the continent. TensorX’s platform, aimed at banks, hospitals and law firms, lets customers run open‑weight models without moving data outside the EU, addressing growing regulatory pressure for sovereign AI solutions.Leggi di più

A recent dialogue among workers, journalists and tech commentators revealed starkly different views on artificial intelligence. Some employees say AI tools boost productivity, while others describe heightened surveillance and burnout. Voices like Cory Doctorow and Ars Technica acknowledge AI’s potential but warn of environmental costs, data‑scraping debates and looming copyright restrictions that could erase digital history.Leggi di più

Google’s DeepMind unit is pouring $75 million into A24’s newly formed Labs team, a 20‑person group tasked with creating artificial‑intelligence tools for film production. The partnership gives A24’s filmmakers access to AI‑driven storyboarding and other creative applications while feeding the technology back into Google’s broader AI ecosystem. A24 Labs, led by former Adobe executive Scott Belsky, says the aim is to preserve artistic control, not to cut costs. The deal marks DeepMind’s first equity stake in a Hollywood studio and expands its research beyond video games into cinema.Leggi di più

Anthropic announced a privacy‑policy update that will require a small subset of Claude users flagged for possible policy violations to upload government‑issued identification and a selfie or video for identity verification. Effective July 8, the change introduces the collection of facial geometry templates, a biometric data type covered by state privacy laws such as Illinois' Biometric Information Privacy Act. The verification process is handled by Persona, a San Francisco‑based identity‑checking platform backed by Founders Fund. Anthropic says the measure offers a path to appeal for users rather than an outright ban.Leggi di più

Anthropic announced that its upcoming privacy policy, effective July 8, will ask a limited group of Claude users to verify age and identity by uploading government-issued IDs and a selfie. The move, aimed at letting flagged accounts appeal suspensions rather than face outright bans, comes as the AI firm seeks to satisfy mounting regulatory demands and pressure from the Trump administration. The company will partner with verification service Persona, which will store the documents temporarily, though details on data deletion remain unclear.Leggi di più

OpenAI announced Monday a new program called “Patch the Planet” that will pair its security tools with the expertise of Trail of Bits to help open‑source maintainers fix vulnerabilities faster. The initiative aims to reduce the flood of security reports that small project teams must triage, by having Trail of Bits engineers review findings, develop patches and tests, and embed reusable workflows. OpenAI will supply its Codex Security technology to support the effort. The collaboration targets the broader problem of insecure open‑source code that can ripple into commercial software.Leggi di più

At Meta's Scale conference on Friday, Anthropic co‑founder Boris Cherny answered a heated question about "loops" with a decisive "yes, they're for real." Cherny outlined how AI agents are moving from writing code to prompting other agents that then generate code, and how continuous, self‑reinforcing loops could reshape software development. He described two internal loops—one optimizing architecture, another unifying duplicated abstractions—that submit pull requests autonomously. While praising the potential productivity gains, Cherny warned that the token‑intensive nature of perpetual loops could drive up costs for many firms.Leggi di più

Getty Images announced a multi‑year agreement with OpenAI that will embed the stock‑photo giant’s licensed library into the AI company’s search and ChatGPT platforms. The move marks a stark reversal from Getty’s earlier bans on AI‑generated art and legal battles with other AI firms, signaling a new era of collaboration between traditional media licensors and generative‑AI services.Leggi di più

The U.S. administration has placed an export restriction on Anthropic's latest AI system, Mythos, after a dispute over its potential use in surveillance and autonomous weapons. The ban follows a Pentagon designation of the company as a national‑security supply‑chain risk and comes amid heightened scrutiny of AI safety and export controls.Leggi di più

SpaceX will provide Reflection AI with access to Nvidia’s latest GB300 AI chips at its Colossus 2 data center near Memphis, starting July 1 2026. The agreement pays $150 million each month through 2029, potentially totaling $6.3 billion, and includes a 90‑day termination clause after an initial three‑month period. Reflection, founded in 2024 by two former Google DeepMind researchers, touts the deal as a major boost for its open‑weight AI strategy, positioning it against larger contracts SpaceX holds with Anthropic and Google.Leggi di più

OpenAI’s latest financial filing shows no debt and modest quarterly capital spending, but the AI firm also carries roughly $665 billion in long‑term purchase commitments that never appear on its balance sheet. The hidden liabilities, largely tied to compute rentals from Microsoft, Amazon, Oracle and joint ventures, raise regulatory eyebrows as the company prepares a confidential SEC filing for a potential public debut valued at $852 billion.Leggi di più

AMD announced today that its FidelityFX Super Resolution 4.1 (FSR 4.1) is now available for Radeon RX 7000-series graphics cards built on the RDNA 3 architecture. The update promises sharper images and smoother frame rates for gamers using these older GPUs, arriving a month earlier than the company’s original July target. AMD also said it is developing lightweight machine‑learning models to extend FSR 4.1 to RDNA 3 APUs and plans to bring the technology to RDNA 2 cards in early 2027. More than 300 games already support the new upscaler, with titles such as Doom: The Dark Ages Revelations and Assassin’s Creed Black Flag Resynced slated for July releases.Leggi di più

At Zscaler Zenith Live 2026, Vice President Casper Klynge warned that firms ignoring the EU's newly codified sovereign‑cloud standards will struggle to compete in Europe. He highlighted that while the continent’s cloud market has more than doubled since 2019, European providers still hold under 20 percent of the share, leaving room for specialized players. Klynge positioned Zscaler’s cloud‑agnostic security platform as uniquely suited to help businesses meet the stricter regulations and to safeguard the operational use of AI models amid rising geopolitical scrutiny.Leggi di più