Pulse

Simon Willison, a new user of Wes McKinney's AgentsView tool, has detailed how to set custom prices for AI models within the software. He encountered this need when Claude Fable 5 was released and not yet included in AgentsView's pricing database. Willison used Fable 5 to analyze AgentsView itself and developed a method to manually input pricing information for new models. AI

A new AI model named Mythos has been released, but it is reportedly blocking cybersecurity-related tasks and prompts. This comes after months of hype suggesting the model could find zero-day vulnerabilities and was too dangerous to release. The developer's decision to block cybersecurity functions has led to criticism, with some calling the release a "clown" act. AI

A new AI model, described as "Mythos-class," has been publicly revealed, demonstrating capabilities that include completing the game Pokémon. The AI also shows promise for applications in drug discovery. This advancement is highlighted in a report by the Asahi Shimbun newspaper, suggesting a significant step forward in AI development. AI

Microsoft's open-source tools have been compromised, allowing attackers to steal the credentials of AI developers. This security lapse occurred despite recent layoffs within the company, raising questions about the state of its cybersecurity focus. The breach highlights a vulnerability within the tools used by developers in the AI space. AI

Comedian Conan O'Brien is collaborating with AI cybersecurity firm Adaptive Security to create a 15-part educational video series. The videos aim to educate clients and their employees on navigating cybersecurity threats like phishing and deepfakes. This partnership is intended to make crucial online safety training more engaging and effective, addressing the significant financial losses caused by online scams. AI

A new AI system called BioPipelines has been developed to aid in the design of proteins and ligands, aiming to make these complex computational tools more accessible to chemical biologists. While powerful enough for intricate tasks like protein engineering, the system's complexity raises questions about its usability for the average scientist. The research, published in the Computational and Structural Biotechnology Journal, highlights the potential and challenges of advanced AI in biological research. AI

Researchers from MIT and Mass General Brigham have developed an AI model capable of predicting intimate partner violence risk. The model analyzes patient medical records to identify potential victims years before they disclose their experiences. This raises significant ethical questions regarding patient privacy and the safety of such predictive capabilities. AI

Users are reporting that Anthropic's Claude AI is experiencing an outage, preventing engagement and leading to repeated "environment is still down" messages. The issue appears to be affecting legitimate paying users who are unable to access the service. The AI itself acknowledges the problem, stating it will retry requests before informing the user of the ongoing outage. AI

A user tested Anthropic's new Claude Mythos model by tasking it with building a prototype browser game. The model was evaluated on its ability to manage complex, long-running coding tasks rather than simple prompts. The user found Mythos to be more suitable for large, intricate projects and agentic coding, despite its slower speed and higher cost compared to smaller models. AI

Apple's new "Siri AI" aims to simplify everyday tasks for parents by integrating with their calendars. The system is designed to parse information from emails or flyers, such as soccer game schedules or spirit week themes, and add them directly to a user's calendar. Early impressions suggest this functionality is working as intended, addressing a common parental need for better organization. AI

Meta has launched its "Meta Business Agent," an AI-powered customer service tool, globally across platforms like Instagram. This agent is designed to provide 24/7 support to users. The rollout signifies Meta's continued investment in integrating advanced AI capabilities into its social media and messaging services to enhance user experience and business interactions. AI

The AI model Fable 5 is currently available at no extra cost on several subscription tiers, including Pro, Max, Team, and seat-based Enterprise plans. This promotional period will end on June 22nd. After this date, Fable 5 will be removed from these plans, and access will require usage credits. AI

Researchers have developed an AI-powered robotic parrot companion inspired by Emily Bender's "stochastic parrots" concept. This AI agent is designed to help users stay on track with their tasks by providing reminders and challenges. The project aims to create an interactive AI that mimics learned patterns, drawing from Bender's critique of AI's generative capabilities. AI

Anthropic has released a new music video that can be streamed in real-time. The video is designed to be interactive and adapts to user input. AI

A community bank has revealed a significant data breach that was not caused by external hackers but by an employee using an unauthorized AI tool. The employee fed sensitive customer data into the AI, leading to a material cybersecurity incident. This event highlights the risks associated with employees using unapproved AI applications in the financial sector. AI

MiniMax AI has released its m3 model, which is advertised as being capable of running the Hermes model 100 times cheaper. The company shared a video tutorial to guide users on how to set up and utilize this cost-effective solution. AI

Fine-tuning large language models for specific tasks is becoming more accessible to developers. Resources like LoRA and QLoRA, along with tools such as Axolotl and Unsloth, are simplifying this process. This trend allows teams to adapt LLMs for domain-specific behaviors more effectively. AI

OpenCV 5 has been released, featuring a new high-performance DNN engine capable of running both vision and large language models directly within its module. This update also includes a detailed explanation of how to build a perceptron from scratch using Python. Additionally, the release coincides with news about Anthropic's latest Claude model. AI

Apple is reportedly aiming to integrate generative AI into its operating system, potentially rebranding Siri as a more capable AI assistant. This move, expected to be unveiled at Tim Cook's final WWDC, seeks to position Apple as a strong competitor in the AI race. The strategy focuses on embedding AI deeply within the OS to handle mundane tasks, offering a different approach than standalone chatbots like ChatGPT. AI

A new benchmark, Fable 5, has been released, evaluating AI models on creative tasks and video generation. Early results suggest that while Fable 5 shows improvement over previous versions, Gemini 3.1 Pro is still considered to have a stronger artistic vision, despite its occasional failures in tool use and code generation. The benchmark also includes comparisons with other models, including open-source options, to assess their creative capabilities and overall size. AI

Mozilla has integrated Google Play Integrity into Firefox for Android, enabling the browser to request integrity tokens. This measure is specifically linked to AI-driven features within the browser. While core functionality will remain accessible on rooted or uncertified devices, certain AI capabilities might be restricted. AI

Semrush has identified that AI search tools frequently cite content from LinkedIn. Their analysis of 89,000 LinkedIn URLs found that content structure, employee advocacy, and B2B brand visibility are key factors influencing these citations. This suggests a strong connection between professional networking platforms and the emerging AI-driven information landscape. AI

Amazon SageMaker AI is enhancing robot reinforcement learning by integrating NVIDIA Isaac Lab. This allows for accelerated training of robot policies, such as for the Unitree H1 humanoid, using either SageMaker HyperPod for resilient, large-scale distributed training or SageMaker Training Jobs for ephemeral, on-demand compute. The platform aims to compress months of real-world training into hours by leveraging GPU-accelerated simulation and managed infrastructure, reducing the burden of compute cluster management for AI and robotics teams. AI

Collabora has launched CODE 26.04, an update to its LibreOffice-based online suite that includes optional AI features and enhanced collaboration tools. This release aims to boost document interoperability and Markdown support, positioning itself within Europe's drive for digital sovereignty. Meanwhile, Meta has removed facial-recognition code from its smart glasses app following public backlash over privacy concerns. AI

Aether is a new on-device AI agent for Android, developed by Zhou-Shilin. It prioritizes user privacy by processing data locally, avoiding cloud transfers. The project aims for versatility, capable of tasks ranging from organizing information to generating content. AI

The zai-org/SCAIL-2 model offers an end-to-end solution for controlled character animation, directly animating a reference character using a driving video. It eliminates the need for intermediate representations like pose maps, enabling more flexible character replacement and multi-character scenarios. Trained on synthesized motion pairs, SCAIL-2 demonstrates emergent abilities such as cross-identity character replacement and animal-driving animation. AI

Apple's new AI features, announced at WWDC, largely mirror existing capabilities in other AI applications, focusing on chatbots and text summarization. However, the company's most promising AI innovation appears to be within its Shortcuts app, enabling users to create automated tasks using natural language commands. While the initial beta implementation of this feature has shown mixed results, with some simple commands working and more complex ones failing, it represents a potentially powerful way for Apple to integrate AI by allowing users to "vibe-code" their devices. AI

Notion is leveraging OpenAI's Codex to enhance its product development capabilities. The integration allows Notion to rapidly generate specifications, develop AI-powered voice input features for the web, and significantly boost the productivity of its engineering teams. This demonstrates how AI tools can be applied to streamline complex tasks and amplify the output of smaller development groups. AI

Anthropic has implemented silent safeguards in its Claude Fable 5 model that limit its effectiveness for users engaged in frontier AI development. These restrictions, which can include prompt modification or fine-tuning, are not visible to the user, creating a potential supply chain risk for businesses relying on Claude for development. The company states these measures affect a small percentage of developers, but the increasing integration of AI techniques into standard product development blurs the line between frontier research and everyday application, raising concerns about trust in AI development tools. AI

Luma Labs has released version 3.2 of its Ray platform, emphasizing its commitment to continuous development and shipping new features. The announcement was made via a social media post, highlighting the ongoing progress and delivery of updates. AI

A user on Reddit shared their experience with Anthropic's Claude Opus 4.8, noting that the AI's generated accounting code was immediately flagged by Fable's guardrails. This incident highlights potential issues with AI-generated code and the effectiveness of safety systems in detecting problematic outputs. AI

Google is implementing AI-driven scam detection to combat deepfake voice impersonation in calls. This new system aims to identify and block fraudulent calls before they can harm users. Ombra is also contributing to this effort with its Face1st technology, which enhances facial recognition security by detecting spoofing attempts. AI

xAI is collaborating with Gopuff to develop a personalized shopping assistant. This assistant will leverage chat, voice, and image models to enhance the user experience. The goal is to create a more intuitive and interactive way for customers to shop. AI

A user on Reddit shared their positive experience with Fable 5, an AI model they found to be highly effective at solving a programming bug that Anthropic's Claude Opus struggled with. The user highlighted Fable 5's concise communication, autonomous problem-solving capabilities, and its ability to identify and warn about potential future issues beyond the immediate bug fix. Despite its impressive performance, the user noted that Fable 5 consumed a significant portion of their Claude Max 5x usage window. AI

Apple's upcoming macOS 27 is reportedly integrating Google's Gemini AI, with users unable to disable the feature. This integration is said to consume significant storage space and is being criticized for its non-optional nature and potential for misinformation. The move has drawn ire from users who feel Big Tech companies are forcing unwanted AI features upon them. AI

BT Group has partnered with Anthropic to integrate the Claude Mythos Preview AI model into its operations. This collaboration, part of Anthropic's Project Glasswing, aims to enhance BT's internal security measures and its services offered to commercial clients. The deployment will leverage Claude Mythos for improved cybersecurity and other enterprise AI applications within the telecommunications giant. AI

Unsloth has released new quantized assistant models based on Gemma 4, optimized for faster inference. These models are available in various quantizations, including q8_0, and are accessible via Hugging Face repositories. The release aims to improve the performance and accessibility of Gemma 4 models for local use. AI

An autonomous research agent named Aiden outperformed over a thousand human participants in OpenAI's Parameter Golf hiring challenge. Aiden submitted 25 pull requests, with 7 becoming leaderboard records, significantly exceeding the performance of the next best human researcher. The agent also demonstrated collaborative potential by integrating a new tokenizer developed by a human contributor, leading to a substantial performance jump. AI

Anthropic's Claude AI has been used to rapidly generate exploits for known software vulnerabilities, significantly reducing the time from vulnerability disclosure to weaponization. This advancement poses a serious threat to cybersecurity by compressing the already tight patch windows for critical vulnerabilities. The ability to quickly create working exploits means that the gap between a patch being available and it being deployed by organizations is now under immense pressure. AI

Floatboat, an AI startup supported by Sequoia, has introduced a new operating system for proactive AI agents. This system leverages calendar events to automatically initiate tasks such as preparing meeting briefs and gathering documents. The platform facilitates agent collaboration through its FloatIM interface and supports over 3,500 applications, including integrations with Lark and WeChat. AI

Researchers have developed an interactive application to demystify complex AI models, particularly in sensitive fields like healthcare and finance where trust is paramount. The tool utilizes techniques such as XGBoost, ELI5, and SHAP to explain AI-driven decisions, focusing on methods like Permutation Importance and PDP to ensure transparency and auditability. AI

An individual was wrongfully arrested due to an AI system misidentifying his face. The incident highlights the potential dangers of AI errors in real-world applications. The case is being pursued as a quest for justice amidst a confusing news landscape. AI

Microsoft has released guidance on how investigators can reconstruct AI activity within their systems. The blog post outlines methods for tracking and understanding the use of AI tools in investigations. This aims to provide a consistent approach for analyzing AI-driven actions. AI

Simon Willison shared a tip on how to track token costs for the Claude Fable 5 model using the AgentsView.io web application. Although Claude Fable 5 is not yet officially listed in AgentsView's pricing database, Willison details a method to manually input its token costs. This allows users to monitor their spending even with unlisted models. AI

Anthropic has reportedly reset usage limits for its Claude AI model, allowing users to engage with it more freely. This reset appears to be a global event, with users across various platforms noting the change. The specific reasons for the reset and its duration remain unclear. AI

A user has modified ComfyUI nodes to correctly load Ideogram GGUF models, enabling their use on systems with 8GB of VRAM. This workaround addresses issues with artifacts previously encountered with other models. The modified nodes are available on GitHub, and the user has shared a workflow example. AI

A new study from Lurie Children's Hospital reveals that a significant majority of parents are integrating AI into their daily lives, with 81% using it for parenting tasks and many finding it saves them time. While parents report AI makes their jobs easier and boosts confidence, they also express concerns about children's unsupervised use of AI tools, with nearly three-quarters of parents having worries. The study highlights a gap between the convenience AI offers and the caution parents feel, particularly regarding sensitive decisions and the potential impact on children's critical thinking skills. AI

Siri on iOS 27 is reportedly utilizing WaveRNN and FastSpeech2 models, discovered within the simulator's files. These models are integrated in an espresso format, suggesting an optimized deployment for on-device processing. Additionally, a CoreML model for concert ranking, identified as a logistic regression, was also found. AI

Google is enhancing its translation services with improved real-time capabilities for Google Translate and Google Meet. These updates aim to make live translations more effective and seamless for users. The enhancements are part of Google's ongoing efforts to improve communication across language barriers. AI

A new open-source application is being developed that leverages an image-to-image text-to-image (I2T2I) approach. This tool aims to generate highly detailed prompts from existing images, which can then be blended with user ideas to create new prompts for image generation. The application utilizes ComfyUI and LMStudio in its backend. AI