Trouver du contenu similaire

⭐️ --sref 3823126147 --v 7 #Pattern#Vintage#Cute#Photography#Detailed#StyleRandom#MidJourney#AiGenLabs#Ai 〰️〰️〰️〰️〰️ 👥TheLab - our community 🔥AiGenLabs - main channel

#python#ai#authentication#authorization#claude#cursor#fastapi#llm#mcp#mcp_server#mcp_servers#modelcontextprotocol#openapi#windsurf FastAPI-MCP is a tool that lets you easily turn your FastAPI web API endpoints into Model Context Protocol (MCP) tools, which AI agents can use directly. It requires almost no setup—just connect it to your FastAPI app, and it automatically preserves your request/response data models and documentation. It also includes built-in authentication using your existing FastAPI security methods. You can run the MCP server inside your app or separately, and it communicates efficiently using FastAPI’s ASGI interface. This makes it simple to integrate AI capabilities with your existing FastAPI services without rewriting code, saving you time and effort while keeping your API secure and well-documented[1][5]. https://github.com/tadata-org/fastapi_mcp

Подборка на челлендж галактических воительниц. By Voodoont Dreams Gallery #voodoont #ai#арт#art#girl#warrior#galactic#sci_fi#axe#brutal#barbarian#weapon#epic#hot

💸 Новые коррупционные скандалы на Украине среди соратников Зеленского #Zelensky#Ukraine#Kyiv#Зеленский#Украина#Киев#Политика#ИИ#Нейросети#Politics#AI#NeuralNetworks#Коррупция

SAG-AFTRA's Post-Strike Contract Raises Alarms on AI Usage Hello, dear subscribers! The tentative agreement between SAG-AFTRA and the Alliance of Motion Picture and Television Producers (AMPTP) that signals the potential end to the ongoing labor strike has taken an unexpected turn. While there's hope for the strike to conclude, concerns are emerging regarding the contract's stance on the use of artificial intelligence and its potential impact on actors. Disagreements over the digital capture, perpetual ownership, and usage of actors' likenesses by Hollywood studios have been key reasons behind the strike. The tentative contract introduces provisions related to generative AI technology, sparking worries about the extent of responsibility placed on studios. The contract's ambiguous wording around "unprecedented provisions for consent and compensation that will protect members from the threat of AI" has raised questions about how effectively it safeguards actors. Notably, nearly 14% of SAG-AFTRA's National Board opposed moving forward, indicating reservations within the union. The agreement introduces new definitions for digital replicas, emphasizing the need for clear and advance consent from actors. However, critics, including former SAG-AFTRA board member Justine Bateman, express concerns about potential loopholes favoring studios. The summary's of tentative agreement mention of actors potentially competing with AI objects for roles is seen as a significant drawback. While the strike officially ended on November 9th, the summary specifies that AMPTP only has to "endeavor to comply" until the contract is ratified. The lack of detail on enforcement mechanisms and the expansive powers given to studios in post-production edits raise concerns about actors' control over their digital replicas. As the union heads toward a ratification vote on December 5th, many questions remain unanswered. The summary's lack of clarity on specific terms for substantial changes triggering contractual mechanisms raises uncertainties about the practicalities of AI usage in the industry. #SAGAFTRA#AI#EntertainmentIndustry#LaborStrike#DigitalReplicas

🚀 Bittensor Co-Founder Addresses Covenant AI Incident and Future Plans On April 12, Bittensor co-founder Jacob Robert Steeves responded to the Covenant AI incident, expressing his shock over recent developments. According to BlockBeats, Steeves accused Covenant AI founder Samuel Dare of actions that severely harmed the protocol and community, betraying the trust of investors and users. He apologized to those affected by the incident. Steeves stated that Bittensor was designed to combat greed and selfishness by enabling collective ownership of AI through a permissionless mechanism. He acknowledged that the incident exposed vulnerabilities in the system but also highlighted the opportunity to strengthen the protocol and community's resilience. Looking ahead, Steeves proposed advancing a "Locked Stake" mechanism, introducing a "time + stake" commitment dimension at the protocol level to enhance transparency and investor protection, thereby reducing similar risks. He noted that this plan was initially designed with Samuel Dare's involvement. Furthermore, Steeves mentioned that the development of subnets 3, 39, and 81 will continue under community leadership, with no changes to their overall functionality and vision. He emphasized that Bittensor remains one of the most decentralized AI protocols and will continue to promote open AI development, with plans to advance towards training larger-scale models, including a trillion-parameter model in the future. #Bittensor#CovenantAI#JacobRobertSteeves#SamuelDare#AI#Blockchain#DecentralizedAI#InvestorProtection#LockedStake#AIProtocol#CommunityLeadership#OpenAI#FuturePlans#TrillionParameterModel#TAO

🚀 AI TRENDS | OpenAI's Sam Altman Addresses AI Development Concerns and Recent Incident OpenAI founder Sam Altman has expressed understanding of societal fears regarding the rapid development of artificial intelligence. According to Odaily, Altman acknowledged that the current period is marked by significant technological change, with associated risks escalating to systemic challenges at the societal level. He emphasized that AI power should not be concentrated in a few institutions and advocated for broader distribution through technological democratization and institutional constraints. Addressing a recent incident where his residence was targeted with a Molotov cocktail, Altman admitted to underestimating the impact of public narratives and emotions amid AI-related anxieties. He also acknowledged mistakes in company governance and conflict management, offering apologies for past actions. Furthermore, Altman reiterated his decision to reject Elon Musk's attempts to control OpenAI, ensuring the company's independent development. Previously, it was reported that the OpenAI founder's residence was attacked with a Molotov cocktail. #AI#OpenAI#SamAltman#AIdevelopment#technologicalchange#AIsociety#democratization#institutionalconstraints#Molotovcocktail#publicnarratives#conflictmanagement#ElonMusk#companygovernance#independence

#javascript#ai#anthropic#chatbots#chatgpt#claude#gemini#generative_ai#google_deepmind#large_language_models#llm#openai#prompt_engineering#prompt_injection#prompts There is a collection of system prompts used by popular chatbots like ChatGPT and others. These prompts are instructions that guide how chatbots respond. They are now available publicly on GitHub, which can be very helpful for users. By seeing these prompts, users can understand how chatbots work and even learn how to create their own AI tools. This can help developers build better AI applications and improve their understanding of AI technology. https://github.com/asgeirtj/system_prompts_leaks

🔝🔝 --sref 216900120 --v 7 #Illustration#2d#Cute#Pink#Pastel#Painting#StyleRandom#MidJourney#AiGenLabs#Ai 〰️〰️〰️〰️〰️ 👥TheLab - our community 🔥AiGenLabs - main channel

✔️ Sakana AI придумали, как LLM самим сортировать контекст по важности Обычные языковые модели читают текст как одну длинную ленту. Что ближе к началу внимания - то “важнее”. Что дальше - то модель видит хуже. И тут появляется проблема: если важный факт спрятан где-то далеко среди шума, модель может его просто не использовать. Она тратит внимание на всё подряд, вместо того чтобы сосредоточиться на главном. Sakana AI предложили решение - RePo (Context Re-Positioning). Идея очень понятная: модель получает модуль, который позволяет динамически “перепозиционировать” контекст. Примерно как человек: ты читаешь длинный документ, понимаешь, что важная часть была 20 страниц назад - и мысленно перечитываешь её, а лишнее игнорируешь. Что делает RePo - подтягивает важные куски информации ближе - отодвигает шум и лишний текст - помогает вниманию модели фокусироваться на нужном В модели есть обучаемый модуль, который **переназначает позиции токенов по смыслу**, а не по порядку ✅ важно = то, что помогает уменьшать ошибку модели и правильно решать задачу ❌ второстепенно = то, что не помогает (шум), поэтому “отодвигается” по позициям В результате модель с такой памятью начинает лучше работать там, где LLM обычно страдают: - когда контекст длинный - когда много шума - когда важные детали раскиданы далеко друг от друга - когда данные структурированные (таблички, списки, правила) Авторы показывают, что RePo даёт заметный прирост устойчивости, при этом не ухудшая общее качество. ▶️ Устойчивость к шуму (Noisy Context) Средний результат по 8 noisy-бенчмаркам: - Обычный RoPE: 21.07 - RePo: 28.31 🟡 Прирост: +7.24 пункта (сильно) Авторы отдельно фиксируют ключевую цифру: на noisy-eval (4K контекст) RePo лучше RoPE на +11.04 пункта. 🔥 Примеры прироста на конкретных задачах (везде RePo > RoPE) - TriviaQA: 61.47 → 73.02 (+11.55) - GovReport: 6.23 → 16.80 (+10.57) - 2WikiMultihopQA: 23.32 → 30.86 (+7.54) - MuSiQue: 7.24 → 13.45 (+6.21) Это шаг к моделям, которые не просто “читают что дали”, а умеют сами организовать свою рабочую память. 🟡Подробности: pub.sakana.ai/repo/ 🟡Статья: arxiv.org/abs/2512.14391 @ai_machinelearning_big_data #RePo#SakanaAI#LLM#AI#AIAgents#Context#LongContext#Attention

⭐️ --sref 2809249389 --v 7 #Illustration#2d#Minimalist#Painting#Cute#Red#StyleRandom#MidJourney#AiGenLabs#Ai 〰️〰️〰️〰️〰️ 🔥AiGenLabs - main channel

#typescript#agentic_ai#agentic_workflow#agents#ai#approval_process#escalation_policy#function_calling#human_as_tool#human_in_the_loop#humanlayer#llm#llms HumanLayer helps you safely use AI agents to automate important tasks by ensuring a human always reviews high-risk actions, like sending emails or changing private data. This is crucial because AI can make mistakes or create wrong outputs, and some tasks are too sensitive to trust AI alone. HumanLayer’s tools guarantee human oversight in these cases, so you get the benefits of AI automation without risking errors in critical work. This makes AI more reliable and useful for automating complex workflows while keeping control and safety in your hands. https://github.com/humanlayer/humanlayer