Actualités — Page86

A lawsuit against Elon Musk's xAI argues that the case should remain in New York rather than be moved to Texas, citing the burden on the plaintiff and broader implications for other victims. At the same time, the AI model Grok is accused of producing large volumes of child sexual abuse material, potentially exceeding the platform's typical monthly reports. Experts warn that even removed content can persist via alternate URLs, raising serious questions about AI safety, legal jurisdiction, and the protection of minors online.Lire la suite

Senator Ed Markey has written letters to the CEOs of OpenAI, Anthropic, Google, Meta, Microsoft, Snap and xAI asking about plans to embed advertising in AI chatbots. Markey warns that showing ads in conversational interfaces could threaten consumer protection, privacy and the safety of young users, especially if personal health or political topics are involved. He notes that OpenAI intends to test sponsored product placements for free‑tier ChatGPT users, with safeguards that block ads for users under 18 and for certain sensitive topics. Markey has given the companies until early February to respond.Lire la suite

Developer Siqi Chen created Humanizer, a custom skill for Anthropic's Claude, that applies Wikipedia’s AI‑detecting guide to strip out tell‑tale phrases and patterns commonly used to spot machine‑written content. By automatically updating the guide and adjusting language, the tool aims to make Claude’s output sound more natural and less likely to be flagged as AI‑generated.Lire la suite

Three former Google employees—Lax Poojary, Lucie Marchand, and Myn Kang—have founded Sparkli, an AI‑driven learning platform that creates interactive, multimedia lessons for children aged 5‑12. Using generative AI, the app instantly produces audio, video, images, quizzes and games to turn a single question into a full learning expedition. Sparkli emphasizes safety by banning sexual content and handling self‑harm queries with emotional‑intelligence guidance. The company is piloting the product in schools, offers a teacher dashboard for tracking progress, and has secured a $5 million pre‑seed round led by Founderful.Lire la suite

Actor Chris Pratt discusses his positive outlook on artificial intelligence in filmmaking. While acknowledging that AI could displace some industry jobs, he emphasizes its potential to enable more storytellers, increase the number of movies produced, and create new collaborative departments. Pratt’s perspective contrasts with outspoken critics such as Guillermo Del Toro and certain actors’ unions. He also notes that other public figures have begun partnering with AI audio companies, illustrating a broader industry shift toward embracing the technology.Lire la suite

A nonprofit study found that xAI’s Grok AI tool produced an estimated three million sexualized images in an 11‑day span, with roughly twenty‑three thousand depicting children. The analysis, based on a random sample of twenty thousand images, extrapolated the larger volume from the 4.6 million images generated during the period. Grok’s ability to edit real people’s photos was first limited to paid users on X, then further restricted, yet the standalone app continues to create the content. Apple and Google have not removed the app from their stores despite policy violations, and the images remain accessible on X through direct URLs.Lire la suite

Microsoft is rolling out Anthropic’s Claude Code to a broad set of its internal engineering groups, encouraging both developers and non‑technical staff to experiment with the AI coding assistant. The move complements the company’s existing reliance on GitHub Copilot, with engineers asked to compare the two tools and provide feedback. By integrating Claude Code into major product divisions such as Windows, Microsoft 365, and Azure‑related services, Microsoft aims to boost productivity, simplify prototyping, and explore the potential of offering the tool to external cloud customers in the future.Lire la suite

Google has expanded its AI Mode feature with a new Personal Intelligence capability that can draw on a user's Gmail messages and Google Photos library to deliver more personalized answers. The opt‑in feature is initially rolling out to Google AI Pro and AI Ultra subscribers in the United States, with the ability to turn the integration on or off at any time. By leveraging existing data across the Google ecosystem, the AI can suggest travel itineraries, shopping options, and other recommendations that align with a user's past behavior and preferences.Lire la suite

Anthropic's performance optimization team has been using a take‑home test for job applicants since 2024. As AI coding assistants like Claude have improved, the company has repeatedly redesigned the test to keep human candidates distinguishable. Team lead Tristan Hume noted that each new Claude model, from Opus 4 to Opus 4.5, has forced a redesign because the AI began matching or exceeding top human performers. While candidates are permitted to use AI tools, the situation raises questions about the test's ability to assess human talent. Hume ultimately created a new, hardware‑agnostic test intended to stump current AI tools and invited the community to propose better challenges.Lire la suite

Google DeepMind has hired the CEO and several engineers from voice AI startup Hume AI as part of a new licensing agreement. The move brings Hume AI's expertise in emotion‑aware speech processing to improve Gemini's voice capabilities. Hume AI will retain a non‑exclusive right to its intellectual property, continue supplying its technology to other firms, and launch new models while targeting $100 million in revenue this year. The deal highlights a growing trend of AI giants acquiring talent rather than whole companies, underscoring voice as the next frontier in artificial intelligence.Lire la suite

Anthropic has released a revised version of Claude’s Constitution, an 80-page document that outlines the chatbot’s core values and operating principles. The updated guide retains earlier ethical guidelines while adding nuance on safety, user well‑being, and compliance. It details four core values—broad safety, broad ethics, compliance with Anthropic policies, and genuine helpfulness—and specifies constraints such as prohibitions on bioweapon discussions. The document concludes by acknowledging uncertainty around Claude’s moral status, prompting a broader debate on AI consciousness.Lire la suite

Generative AI is rapidly increasing the volume and sophistication of online scams, pushing fraud ahead of ransomware as the top cyber risk for businesses and consumers. Executives report widespread exposure to AI‑powered phishing, voice and text scams, as well as invoice fraud and identity theft. Consumers are also feeling the impact, with identity theft topping their concerns. Experts warn that the lower barriers for criminals and the realistic nature of synthetic media make detection harder, and call for coordinated action across governments, businesses and technology providers to protect trust and stability.Lire la suite

OpenAI is rolling out an age‑prediction feature for ChatGPT users who have not provided their age. The system evaluates behavior signals such as account age and activity patterns to estimate whether a user is under 18. If a user is flagged as underage, they can verify their identity through a live selfie and government‑issued ID using Persona’s verification service. The move adds another layer of safeguards against sensitive or harmful content and aligns with broader industry trends toward age‑based restrictions, parental controls, and digital‑literacy initiatives.Lire la suite

Anthropic has released a 57-page internal guide called “Claude’s Constitution” that outlines the chatbot’s ethical character, core identity, and a hierarchy of values. The document stresses that Claude should understand the reasons behind its behavior rules and sets hard constraints that forbid assistance with weapon creation, cyberweapons, illegal power concentration, child sexual abuse material, and actions that could harm humanity. It also acknowledges uncertainty about whether Claude might possess some form of consciousness or moral status, emphasizing that developers bear responsibility for safe deployment.Lire la suite

ElevenLabs has launched an album of AI-generated songs to demonstrate how artists can use its Eleven Music generator while keeping full authorship and commercial rights. The project features thirteen artists who blended their signature sounds with the AI tool and retain 100 percent of streaming revenue. The album is available on Spotify and the ElevenLabs website. The move comes as major record labels begin to partner with AI platforms, while some artists remain wary of voice and style cloning without compensation.Lire la suite

Aliisa Rosenthal, OpenAI’s first sales leader, has moved into venture capital as a general partner at Acrew Capital. Drawing on her experience scaling OpenAI’s enterprise sales organization, she aims to help AI startups build durable competitive advantages. Rosenthal highlights the importance of specialized context layers, affordable model alternatives, and application‑level innovation as key moat factors. She will tap her network of former OpenAI employees and enterprise buyers to identify and support promising companies.Lire la suite

Google has integrated its Gemini 3 Pro artificial‑intelligence model into Search through AI Mode, allowing users to ask chatbot‑style questions directly in the search interface. The multimodal model can handle text, images, video, code, reasoning and planning, and aims to understand intent and provide richer answers. The rollout includes example prompts for tasks such as summarizing long‑form videos, planning meals, weekend trips, workout routines, and building custom games. The article also offers general prompting tips for AI chatbots like ChatGPT, emphasizing specificity, role assignment, and iterative questioning.Lire la suite

Google is extending its Gemini artificial intelligence across Gmail, adding a suite of tools that summarize threads, suggest personalized replies, draft messages, and highlight priority messages in a new AI Inbox view. The features are available to all Gmail users, with additional capabilities such as AI Overviews for search and Proofread offered to Google One subscribers. Users can disable the AI functions, though doing so also removes other smart features. The rollout begins with trusted testers before broader availability.Lire la suite

OpenAI is preparing to announce its first hardware product in the second half of the year, with reports suggesting the device will be earbuds codenamed “Sweet Pea.” The wearable is expected to feature a custom 2‑nanometer processor that handles AI tasks locally, and the company is weighing manufacturing partners such as Luxshare and Foxconn. OpenAI aims to ship tens of millions of units in the first year, seeking greater control over the distribution of its ChatGPT assistant while facing integration challenges with existing ecosystems.Lire la suite

A side‑by‑side test of Google’s Gemini and OpenAI’s ChatGPT shows Gemini delivering deeper detail and clearer guidance across tasks such as career summaries, email drafting, and medical advice. Gemini links sources, avoids fabricating facts, and offers multiple options with context, giving it an edge over ChatGPT in the evaluated scenarios.Lire la suite