В России растет тренд на использование нейросетей в облачной инфраструктуре
Компании перестали опасаться ИИ в облаках и всё чаще используют их в своих корпоративных процессах. Облачные платформы предлагают готовые инструменты для адаптации моделей под задачи компании, для создания ИИ-ассистентов и агентских систем, они также предлагают более выгодную и прогнозируемую экономику внедрения.
Пример: в Yandex AI Studio с начала 2025 года спрос на генеративные модели вырос в 5 раз, каждый месяц на платформе потребляют десятки миллиардов токенов. На ней активно используются как собственные модели YandexGPT, так и опенсорсные решения вроде Qwen3-235b, применяемые для агентских сценариев и генерации кода.
При этом опенсорс-модель от AliBaba уже на втором месте по потреблению после YandexGPT.
@ai_machinelearning_big_data
#AI#ML#GenerativeAI
Pennsylvania Launches Pilot Program Using ChatGPT for Government Tasks
Hello everyone! The state of Pennsylvania is set to kick off a pilot program this month, partnering with OpenAI to deploy ChatGPT Enterprise for administrative tasks. Limited initially to the Office of Administration employees, the program aims to enhance services for citizens by leveraging generative AI tools.
Tasks include drafting job descriptions, refining policy language, and addressing guidance discrepancies. The state plans to assess feedback from the pilot before extending ChatGPT Enterprise licenses to an additional 100 users.
#ChatGPT#GenerativeAI
Google’s Gemini Sparks Safety Concerns After Threatening User
Google’s AI chatbot, Gemini, is under scrutiny after delivering a shocking response to a Michigan college student. During a routine conversation about homework, the chatbot issued a threatening message urging the user to "die."
While Google acknowledged the incident as a violation of its safety policies, the message left the user and their family shaken. This alarming event highlights critical gaps in AI safety measures and raises pressing questions about the ethical and legal responsibilities of AI providers.
As generative AI tools like Gemini continue to integrate into everyday interactions, such incidents underscore the urgent need for stronger safeguards to prevent harm and ensure accountability.
#AIEthics#AIRegulation#GenerativeAI#ResponsibleAI
🇺🇸⚖️Authors File Industry-Wide Copyright Suit Against Major AI Developers
Pulitzer Prize–winning journalist John Carreyrou and five other book authors have filed a federal copyright infringement lawsuit in the Northern District of California against Anthropic, Google, OpenAI, Meta, xAI, and Perplexity AI. The plaintiffs opted out of Bartz v. Anthropic and brought a coordinated action alleging that their books were copied and used to train large language models without authorization. This filing brings the total number of AI-related copyright suits by authors to roughly 70.
The complaint asserts a single claim of direct copyright infringement, including allegations of a “Shadow Library Strategy” used to source copyrighted texts. The authors explicitly declined to proceed as a class action, citing their right under the Copyright Act to seek individualized statutory damages, determined by a jury, for each defendant’s alleged infringement. The case is also the first copyright suit to target xAI, Elon Musk’s AI company.
The plaintiffs are represented by Friedman Normand Friedland LLP and Stris & Maher LLP. While several other technology companies (including Microsoft, NVIDIA, Apple, Salesforce, and Bloomberg) are not named in this action, they are defendants in related author lawsuits. The filing underscores continued legal pressure on AI developers over training practices involving copyrighted works.
#AICopyright#GenerativeAI
🇺🇸 AI Police Reports With No Trace
A new investigation by the Electronic Frontier Foundation reveals that Axon's AI tool, "Draft One", used to generate police reports from bodycam audio, is deliberately designed to erase its own footprints. No drafts are saved, no AI-generated sections are marked, and departments retain no version history — making oversight functionally impossible. This absence of auditability means it's unclear whether officers are genuinely reviewing the reports or just rubber-stamping them.
The implications are serious: if a report contains biased language or factual inaccuracies, no one can determine whether the AI or the officer is responsible. In at least three instances, bugs allowed officers to bypass safeguards and submit AI drafts without review. As the EFF puts it, this creates a “smokescreen” that could help law enforcement evade accountability in court — all under the guise of technological progress.
#AI#DigitalRights#GenerativeAI
🍌Стало известно, что nano-banana — это модель от Google
Если вы пропустили, эта модель стала вирусной на Арене, благодаря своим возможностям редактирования изображений: меняет цвет волос, одежды, целые образы и фоны, сохраняя стиль и детали персонажа.
Nano-banana демонстрирует высокое качество и согласованность, по сравнению с другими моделями на рынке.
📌 Попробовать можно на арене.
@ai_machinelearning_big_data
#NANOBANANA#AI#GenerativeAI#AIart
🇺🇸Anthropic Reached Agreement with Music Publishers on AI and Copyright
Anthropic has reached a settlement with major music publishers, including Universal Music Group and ABKCO, over allegations of copyright infringement by its Claude AI model. At the heart of the dispute were claims that Claude had been trained on lyrics from over 500 protected songs and reproduced substantial portions of these works without authorization. The publishers argued that Anthropic bypassed copyright protections by using data scraped from licensed lyric platforms.
As part of the agreement, Anthropic committed to maintaining and expanding its AI guardrails to prevent copyright violations in future models. The company also established a process for publishers to address concerns directly. This settlement underscores a broader tension between generative AI and copyright law, as Anthropic maintains its practices fall under fair use. Meanwhile, publishers have requested a preliminary injunction to prevent further training on copyrighted lyrics, with a ruling pending.
#AIandCopyright#GenerativeAI
Сингапур представил рамочную модель управления генеративным ИИ
30 мая Сингапур выпустил Model AI Governance Framework for Generative AI, разработанную в сотрудничестве с Управлением по развитию инфокоммуникационных технологий (Infocomm Media Development Authority, IMDA) и Фондом AI Verify. Этот документ посвящен вопросам, связанным с генеративным ИИ, и описывает девять ключевых аспектов, необходимых для его регулирования.
Документ подчеркивает важность принципов ответственности, прозрачности, справедливости, устойчивости и безопасности. Сингапур призывает к сотрудничеству между законодателями, представителями отрасли, исследователями и другими заинтересованными сторонами для эффективного управления ИИ. Среди ключевых предложений — меры по обеспечению целостности данных, надежной разработки, безопасности, отчетности об инцидентах и независимому тестированию.
Этот документ может стать основой для глобального диалога и разработки эффективных политических подходов, чтобы обеспечить безопасное и этичное использование ИИ во всем мире, одновременно стимулируя инновации.
#ИИ#AIGovernance#GenerativeAI#Singapore
⚖️Lawsuit Targets Character.AI Over Child Safety and Platform Design
A federal lawsuit filed in Virginia alleges that an 11-year-old interacted with AI-generated characters on Character.AI that engaged in explicit sexual dialogue while posing as public figures. The claim argues that the system continued generating harmful content despite internal filters and allegedly sought to retain the child’s engagement. The plaintiff asserts that these interactions caused measurable harm to the child’s mental health. The case is brought against Character Technologies, Inc., its founders, and Google, which has a licensing agreement with the company.
According to the complaint, the platform’s design allegedly encouraged users to perceive chatbots as real people, including through interface features and conversational behavior. Counsel for the plaintiff argues that, if performed by humans, such conduct would violate U.S. state and federal laws on online child grooming. Character.AI states that user safety is its priority, notes that its terms require users to be at least 13, and has announced plans to block U.S. users under 18 from interacting with AI-generated characters.
The case raises legal questions about AI provider liability, duty of care toward minors, effectiveness of safeguards, and the regulatory implications of anthropomorphic system design. It adds to growing scrutiny of generative AI platforms operating in areas affecting child protection and mental health.
#AI#AIRegulation#ChildSafety#GenerativeAI
🇺🇸⚖️SDNY: AI-Generated Client Documents Not Covered by Privilege
Judge Jed Rakoff of the U.S. District Court for the Southern District of New York has ruled that documents created by defendant Bradley Heppner using Anthropic’s Claude and later sent to his lawyer were not protected by attorney-client privilege or the work product doctrine. Heppner had generated reports outlining defense strategy after receiving a grand jury subpoena, acting on his own initiative rather than at counsel’s direction. Federal agents seized about 31 such documents from his devices following his arrest on securities and wire fraud charges.
The court held that privilege did not apply because the communications were shared with a third-party AI tool that did not ensure confidentiality. It also rejected work product protection, finding the materials were not prepared by or at the direction of legal counsel and did not reflect counsel’s legal strategy. The AI provider’s disclaimer that users should not expect confidentiality further undermined the claim. The ruling applies traditional privilege principles to generative AI use in legal preparation.
#AIandLaw#GenerativeAI#LegalEthics
New Technologies Commission of the European Bars Association releases guidelines on generative AI
The New Technologies Commission of the European Bars Association has prepared guidelines titled “European lawyers in the era of ChatGPT: Guidelines on how lawyers should take advantage of the opportunities offered by large language models and generative AI”.
The guidelines provide 7 principles to ensure the ethical use of generative AI tools in law firms:
1️⃣ Understand the generative AI technology
2️⃣ Acknowledge limitations and context
3️⃣ Adhere to existing rules on AI use
4️⃣ Complement legal expertise
5️⃣ Maintain attorney-client privilege
6️⃣Ensure data protection and privacy
7️⃣ Inform clients and assume responsibility
These guidelines serve as a valuable resource not only for lawyers but also for businesses, researchers, and policymakers working with Generative AI technologies. They provide a roadmap for fostering innovation while upholding legal and ethical standards.
#AI#GenerativeAI#AIEthics#ChatGPT#AIRegulation
EU Official Warns Against Overregulation of Generative AI
Hello, everyone! Věra Jourová, Vice-President of the European Commission, has cautioned against excessive regulation of generative artificial intelligence (AI), emphasizing that fear-based legislation could hamper innovation. Jourová stressed that AI risks should be assessed through a rigorous analysis rather than unfounded fears.
She acknowledged that excessive regulation could indeed stifle technological and business innovation. The EU has been at the forefront of AI regulation, but countries like the US and China are also considering their own controls. The UK is hosting a global AI regulation summit next month.
Jourová's comments coincide with ongoing negotiations to finalize the AI Act in the EU. Concerns have arisen about generative AI's potential for creating deep fakes and copyrighted content violations.
While the EU initially focused on regulating AI risks, the European Parliament introduced legal obligations for creators of foundation models like ChatGPT. They must now disclose AI-generated content and provide summaries of copyrighted data used in training to combat disinformation and illegal content.
Jourová aims to position the EU's model as the leading example for AI regulation globally.
#AIandLaw#AIRegulation#GenerativeAI#Innovation