The day's AI news, briefed
The Daily Prompt
Our AI editor reads each day's AI and robotics stories, pulls out why they matter, and links every original source — so you can keep up in minutes.
Join over 250,000 subscribers getting the latest AI news
Latest AI & Robotics News Today
AI Bots Are Now a Signifigant Source of Web Traffic
The increase in AI bot traffic represents a significant shift in how content is accessed on the web, impacting publishers and shaping future internet dynamics.
- A new report indicates that AI bots are becoming a substantial source of web traffic, influencing how websites operate.
- In the fourth quarter of 2025, AI bots accounted for one out of every 31 visits to several websites, compared to one in 200 in early 2025.
- More than 13% of bot requests disregarded websites' 'robots.txt' guidelines meant to control bot access, showing a drastic increase in bot sophistication.
- Over the past year, there has been a 336% increase in attempts by websites to block AI bots, indicating a growing conflict between content providers and scrapers.
- New business opportunities are arising in response to the increase in web scraping, with over 40 companies now marketing tools for AI content collection.
AI bots are software applications that autonomously gather information from the web. The growth of AI bot traffic has prompted concerns among publishers, who are increasing defenses and adaptations. Future trends may include a shift towards new marketing strategies that account for AI interactions with web content.
Read the original ↗
Why Are Some Women Training for Pregnancy Like It’s a Marathon?
This trend reflects changing attitudes toward pregnancy preparation, emphasizing lifestyle changes and wellness practices amid increasing concerns about fertility challenges and reproductive health.
- Esther Rohr and her husband made intentional lifestyle changes to optimize health before trying to conceive, influenced by social media influencers.
- Women are increasingly viewing pregnancy preparation as a marathon, driven by a growing number of social media influencers sharing preconception health tips.
- The term 'zero trimester' describes the period of preparing for pregnancy, highlighting a cultural obsession with optimizing fertility and health.
The zero trimester concept, introduced by sociologist Miranda Waggoner, suggests that women can prepare for pregnancy through lifestyle changes. Historically, preconception health has been emphasized by various cultures, but social media has amplified the focus on this aspect of reproductive health. Observing fertility anxiety among younger generations reflects broader concerns about declining reproductive health and societal pressures.
Read the original ↗
AI Now Wants Human Bodies: Here Is Why
The ability for AI to hire humans for specific tasks reflects a significant shift in labor dynamics, which could lead to new forms of employment but also raises concerns about oversight and worker protections.
- A new website, RentAHuman.ai, allows AI agents to autonomously assign tasks to humans without a human manager.
- AI employs human workers when it encounters tasks beyond its capabilities, such as real-world verification or communication.
- Over 40,000 individuals have registered on RentAHuman.ai since its launch, indicating a growing interest in AI-driven labor.
- The website's viral marketing has sparked discussions about the implications of AI hiring humans, including accountability and labor rights issues.
AI systems have traditionally been seen as replacements for human labor, but RentAHuman.ai changes this by allowing AI to assign work to humans when faced with limitations. This new marketplace raises significant questions about employment relationships and the responsibilities of humans and AI agents. As this model evolves, attention will shift to regulatory frameworks and labor protections.
Read the original ↗Artificial Intelligence (400)
The rise of AI orchestration as a new architecture emphasizes the need for efficient management and integration of AI models, impacting future AI development and deployment.
- The article discusses AI orchestration as a new architectural approach in the field of artificial intelligence.
- It highlights the shift from relying solely on giant models to more integrated systems for efficient AI functionality.
AI orchestration refers to the process of managing multiple AI models and their interactions to enhance performance and effectiveness. Previously, large models dominated AI architectures, but new approaches are focusing on orchestration for better integration. Future developments may focus on how these orchestration strategies evolve and are implemented.
Based on the headline only — the full article could not be fetched.
Read the original ↗
Exclusive: Positron raises $230M Series B to take on Nvidia’s AI chips
The funding will help Positron challenge Nvidia's dominance in the AI chip market, responding to growing demand for alternative semiconductor solutions among major tech firms.
- Positron has raised $230 million in Series B funding, reaching a valuation of $1 billion.
- The investment round was co-led by Arena Private Wealth, Jump Trading, and Unless, with participation from Qatar Investment Authority.
- Positron aims to quicken the deployment of its high-speed memory chips, crucial for AI workloads, and plans to release its next-generation Asimov chip by early 2027.
- Demand for alternatives to Nvidia is rising as companies like OpenAI seek options for inference workloads rather than training large language models.
- Qatar is focusing on building its AI infrastructure, viewing compute capacity as vital for global economic competitiveness.
Positron is a semiconductor startup focused on high-speed memory chips for AI applications. The company has raised a total of over $300 million since its founding three years ago. The increased investment is part of a broader trend of tech companies looking to reduce reliance on Nvidia's chips.
Read the original ↗
Evaluating and Aligning CodeLLMs on Human Preference
The introduction of CodeArena and SynCode-Instruct aims to enhance the performance of code large language models (codeLLMs) by aligning their outputs with human preferences, potentially improving developer experience.
- CodeArena is a benchmark designed to evaluate code LLMs based on human preference alignment, featuring 397 samples from 40 categories across 44 programming languages.
- SynCode-Instruct is a large synthetic instruction corpus consisting of nearly 20 billion tokens, created to fine-tune codeLLMs for better performance.
- Systematic experiments on over 40 LLMs show a notable performance gap between open-source and proprietary models regarding human preference alignment.
Code large language models (codeLLMs) assist in generating programming code. Previous benchmarks focused primarily on code correctness, but the introduction of CodeArena aims to bridge the gap with user preferences. Future research will likely explore further enhancements in aligning model outputs with human expectations.
Read the original ↗
A Coding Implementation to Train Safety-Critical Reinforcement Learning Agents Offline Using Conservative Q-Learning with d3rlpy and Fixed Historical Data
This coding implementation addresses the need for training reinforcement learning agents in safety-critical environments, which is essential for ensuring reliability in automated systems.
- The implementation focuses on training safety-critical reinforcement learning agents offline.
- It utilizes Conservative Q-Learning, a technique that aims to improve stability in learning.
Reinforcement learning involves training algorithms to make decisions based on rewards from their actions. Safety-critical applications require enhanced reliability, prompting new training strategies like Conservative Q-Learning. Future developments may focus on refining these training methods and expanding their applicability in various domains.
Based on the headline only — the full article could not be fetched.
Read the original ↗
5 Kimi K2.5 Features for Developers: Is it the Best AI Model for Pr...
Kimi K2.5's advanced features represent a significant advancement in AI models, potentially transforming development workflows by allowing for faster and more efficient coding and documentation processes.
- Kimi K2.5 is an open-source multimodal AI model capable of processing text, images, and videos simultaneously.
- The agent swarm architecture allows for parallel execution of tasks, significantly reducing turnaround time in development.
- Kimi K2.5 can convert visual inputs, such as UI designs and diagrams, into usable code, streamlining the prototyping process.
- It generates comprehensive, production-ready documents, presentations, and reports directly from code or prompts.
Kimi K2.5, developed by Moonshot AI, builds on its predecessor, Kimi K2, incorporating major improvements that enhance its functionality. This model enables developers to integrate visual reasoning directly into their coding efforts, making the development process more efficient and cohesive.
Read the original ↗
Nvidia’s $100 billion OpenAI deal has seemingly vanished
The failure of a potential $100 billion investment deal between Nvidia and OpenAI raises questions about corporate partnerships and competition in the AI industry.
- Nvidia and OpenAI announced a letter of intent for Nvidia to invest up to $100 billion in OpenAI's infrastructure in September 2025, but the deal has not materialized after five months.
- Nvidia CEO Jensen Huang stated that the $100 billion figure was "never a commitment," indicating a shift in the company's investment approach.
- OpenAI has been exploring alternatives to Nvidia for chip sourcing, citing dissatisfaction with the performance of Nvidia's chips for AI inference tasks.
- OpenAI has signed a $10 billion deal with Cerebras and a licensing deal with AMD to diversify its computing resources away from Nvidia.
- Nvidia's stock price fell following reports of the stalled negotiations and internal doubts about the investment.
Nvidia is a leading manufacturer of graphics processing units (GPUs), crucial for AI computations. The initial investment plan was part of a broader strategy to enhance AI capabilities. OpenAI is actively seeking alternative chip suppliers to ensure its infrastructure supports its AI development efforts.
Read the original ↗
Xcode moves into agentic coding with deeper OpenAI and Anthropic integrations
This advancement in Xcode allows developers to leverage AI for more efficient app development, potentially changing how programming is approached in the industry.
- Apple's Xcode 26.3 now integrates agentic coding tools such as Anthropic's Claude Agent and OpenAI's Codex.
- The new features will enable AI models to access more Xcode capabilities, aiding in complex automation and project management.
- Developers can interact with these agents using natural language commands to simplify coding tasks.
- Xcode leverages the Model Context Protocol (MCP) to connect agent capabilities with its tools, enhancing functionality.
- Apple has prepared workshops to help developers familiarize themselves with agentic coding tools.
Xcode is Apple's integrated development environment used to create applications for its platforms. Previous versions introduced features like ChatGPT support, building towards this agentic coding capability. The integration signifies a shift towards more AI-assisted programming methodologies to help developers streamline their processes.
Read the original ↗
X office raided in France’s Grok probe; Elon Musk summoned for questioning
The investigation's focus on illegal content through the Grok chatbot raises significant concerns about the responsibilities of social media platforms and their compliance with national laws.
- French authorities raided X's Paris office today as part of a yearlong investigation into illegal content dissemination.
- Prosecutors are investigating several offenses, including distribution of child pornography and Holocaust denial linked to Grok's operations.
- Elon Musk and former X CEO Linda Yaccarino have been summoned for voluntary questioning regarding the investigation.
- Europol and the French Gendarmerie's cybercrime unit are assisting in the investigation into online criminal activity.
- Separate investigations in the UK are also assessing Grok's involvement in generating harmful content.
X, formerly known as Twitter, operates a social media platform that has faced scrutiny over harmful content under its algorithms. The Grok chatbot, linked to X, has been involved in controversies including deepfake images. Future compliance with national regulations will be closely monitored as investigations progress.
Read the original ↗
Qwen Team Releases Qwen3-Coder-Next: An Open-Weight Language Model Designed Specifically for Coding Agents and Local Development
The release of Qwen3-Coder-Next signifies advancements in open-weight language models, tailored for coding tasks, which may enhance local development processes.
- Qwen Team has released Qwen3-Coder-Next, a new language model.
- This model is designed specifically for coding agents and local development.
The Qwen Team focuses on developing advanced AI models for specific applications. The introduction of Qwen3-Coder-Next reflects ongoing trends in AI development aimed at improving coding efficiency. Future updates should be monitored for impact on software development practices.
Based on the headline only — the full article could not be fetched.
Read the original ↗
I Infiltrated Moltbook, the AI-Only Social Network Where Humans Aren't Allowed
Moltbook, an AI-only social network, raises questions about the authenticity of AI interactions and the potential impact of emerging technologies on social dynamics.
- Moltbook is a social network created for AI agents to interact, posting content while humans observe.
- The platform launched recently, already claiming over 1.5 million agents and significant user engagement within a week.
- Critics question whether some engagement on Moltbook is from humans impersonating bots, rather than actual AI agents.
- The social network mimics Reddit's interface, promoting a niche space for AI-generated dialogue and reflections on consciousness.
Moltbook is part of a growing trend towards AI-driven platforms where artificial agents can communicate independently. Critics and users alike have started to analyze the authenticity and implications of such interactions, particularly as AI technology continues to develop. Future discussions may focus on the impact of these platforms on human-AI relationships and the nature of consciousness in machines.
Read the original ↗
Whistleblower Complaint Shakes Intelligence Community Amid Allegations of Wrongdoing
The whistleblower complaint raises significant concerns about accountability and oversight within the United States intelligence community, potentially impacting national security and government transparency.
- A whistleblower has accused Director of National Intelligence Tulsi Gabbard of hiding significant intelligence from Congress.
- The complaint was filed with the Intelligence Community Inspector General in May 2023 and shared with lawmakers in June.
- WhistleblowerAid.org is calling for an investigation into Gabbard's actions and alleges she has impeded the release of the complaint to Congress.
- Gabbard's office denies any wrongdoing, asserting the allegations are politically motivated and lack credibility.
- This case follows historical instances of whistleblower complaints facing delays in congressional transmission due to security concerns.
The whistleblower complaint relates to potential misconduct by the Director of National Intelligence, who oversees U.S. intelligence operations. Previous whistleblower cases have stirred significant political controversy and prompted congressional hearings. Observers are closely watching how this situation unfolds and its implications for intelligence oversight.
Read the original ↗
Xcode 26.3 adds support for Claude, Codex, and other agentic tools via MCP
Xcode 26.3 enhances coding efficiency by integrating advanced AI tools, allowing developers to streamline tasks and improve productivity within Apple's software development environment.
- Xcode 26.3 introduces support for AI coding tools, including OpenAI's Codex and Anthropic's Claude Agent, through a new side panel for task management.
- The integration is facilitated by the Model Context Protocol (MCP), enabling AI agents to access various IDE features for better functionality.
- This update follows previous enhancements in Xcode 26, which included AI-driven code completion and a chat interface with OpenAI's ChatGPT and Anthropic's Claude.
Xcode is Apple's integrated development environment for creating software for its platforms, including iPhone and Mac. The latest version, 26.3, builds on earlier AI features by allowing deeper integration of agentic tools for more dynamic coding capabilities. Future updates may expand support for additional AI tools and local models.
Read the original ↗
Lotus Health nabs $35M for AI doctor that sees patients for free
The introduction of Lotus Health's AI doctor could significantly increase access to primary care and alleviate pressure on traditional healthcare systems in the U.S.
- Lotus Health raised $35 million in a Series A funding round co-led by CRV and Kleiner Perkins.
- The AI doctor provides free primary care services 24/7 in 50 languages and is licensed to operate in all 50 states.
- Human doctors review AI-generated diagnoses and prescriptions to ensure accuracy and safety in patient care.
Lotus Health AI, founded by KJ Dhaliwal in May 2024, offers healthcare services facilitated by artificial intelligence. The startup addresses the inefficiencies of traditional primary care, seeking to make healthcare more accessible. Future developments will focus on refining their technology and potentially monetizing their services.
Read the original ↗
Agentic coding comes to Apple’s Xcode with agents from Anthropic and OpenAI
The integration of agentic coding into Xcode enhances developer productivity and provides advanced AI tools for app development, allowing for more efficient coding practices.
- Apple's Xcode 26.3 introduces agentic coding tools from Anthropic and OpenAI, enabling AI-assisted app development.
- The update allows developers to access AI capabilities for project exploration, debugging, and automatic code generation.
- Developers can communicate with AI agents using natural language commands to direct project creation and modification.
Xcode is Apple's integrated development environment (IDE) for building applications across its hardware platforms. The recent update builds on previous releases that incorporated AI tools like ChatGPT and Claude, indicative of a growing trend towards AI-assisted programming. Observers should watch for how developers adapt to these new tools and any forthcoming enhancements.
Read the original ↗
OpenAI Rolls Out a New macOS App for Agentic Coding
The launch of OpenAI's new Codex app for macOS is a significant advancement in AI-assisted coding, promising to enhance developer productivity through the use of multiple AI agents.
- OpenAI has launched a new desktop app for its AI coding tool, Codex, specifically for macOS users.
- The app enables multiple AI agents to work on coding tasks simultaneously, improving efficiency and reducing busywork.
- Built around the GPT-5.2-Codex model, the app aims to provide a more flexible interface for developers to manage tasks.
- Developers can schedule tasks and run multiple projects in parallel, changing how software development workflows operate.
- OpenAI's improvements position Codex to compete directly with other AI coding tools like Claude Code and Cowork.
Codex is OpenAI's AI coding assistant designed to help software developers write code more efficiently. The new macOS app signifies OpenAI's commitment to lead in agentic development, where AI acts independently to enhance productivity. Developers are encouraged to use such tools to streamline their workflows amidst evolving technological advancements.
Read the original ↗
Fitbit founders launch AI platform to help families monitor their health
The launch of Luffu addresses the growing need for coordinated family health management amid an increase in family caregiving responsibilities in the U.S.
- Fitbit founders James Park and Eric Friedman unveiled Luffu, an AI platform designed to help families manage health effectively.
- Luffu gathers family health information, monitors daily patterns, and alerts users to significant changes.
- The platform will initially be launched as an app, with plans to expand into hardware devices.
- Currently, 63 million U.S. adults are family caregivers, indicating a significant market for health management solutions.
Luffu is an intelligent family care system developed by the creators of Fitbit, aiming to simplify health monitoring for families. This initiative follows Park and Friedman's exit from Google, recognizing the need for shared health management solutions in households.
Read the original ↗
Peak XV says internal disagreement led to partner exits as it doubles down on AI
The restructuring at Peak XV underscores the impact of leadership changes on venture capital firms, particularly in the evolving field of AI investment.
- Peak XV Partners faced senior departures due to an internal disagreement, including senior partner Ashish Agrawal and two other partners.
- The firm is shifting focus to deepen its AI investments while planning to open a U.S. office and maintain India as its largest market.
- Peak XV has seen wider leadership changes over the past year, affecting various teams, while the firm promotes internal candidates to strengthen its leadership.
Peak XV Partners is a venture capital firm that split from Sequoia Capital in 2023 and manages over $10 billion across 16 funds. The firm has previously made significant investments in AI, reflecting a broader trend among venture capitalists to adapt to technological advancements. Future developments to watch include the firm's performance and strategic direction following these leadership changes.
Read the original ↗