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
Alibaba Qwen Unveils Qwen3-4B-Instruct-2507 and Qwen3-4B-Thinking-2507: Refreshing the Importance of Small Language Models
The introduction of these powerful small language models may redefine efficiency and performance standards for AI applications in various fields, making advanced technologies more accessible.
- Alibaba has launched two new small language models, Qwen3-4B-Instruct-2507 and Qwen3-4B-Thinking-2507, both featuring 4 billion parameters and 256K context support.
- The Qwen3-4B-Instruct-2507 model is optimized for speed and clarity, providing concise responses in over 100 languages for customer support and education.
- The Qwen3-4B-Thinking-2507 model excels in deep reasoning and problem-solving, suitable for technical diagnostics, scientific interpretation, and advanced coding tasks.
Alibaba's Qwen team specializes in developing advanced AI language models. The new models emphasize long-context capabilities and efficient performance on consumer-grade hardware, reflecting a trend towards smaller yet powerful AI solutions. Future developments may include further enhancements in multilingual support and reasoning capabilities.
Read the original ↗
Anthropic revenue tied to two customers as AI pricing war threatens margins
Anthropic's reliance on two major customers for a significant portion of revenue exposes the company to risks as market dynamics shift with competitive pricing pressures.
- Anthropic's revenue run rate has reached $5 billion, heavily reliant on Cursor and GitHub Copilot, which together drive about 25% of its income.
- The launch of OpenAI's GPT-5 introduces lower pricing that threatens Anthropic's premium market position and could pressure its pricing strategy for Claude.
- Despite this, Anthropic holds 42% of the AI coding market, outpacing OpenAI's 21% share, indicating significant growth in enterprise AI usage.
Anthropic, known for its Claude AI assistant, is heavily focused on AI-powered software development, a growing sector that has attracted substantial enterprise interest. The competitive landscape is rapidly evolving as OpenAI introduces lower-cost alternatives, prompting companies to reassess their AI vendor partnerships.
Read the original ↗OpenAI gets caught vibe graphing
Miscalculated graphics in a major product presentation could undermine trust in AI data and impact public perception of OpenAI's credibility and its new GPT-5 model.
- OpenAI's CEO Sam Altman identified a significant error in graph representations during the GPT-5 presentation, referring to it as a 'mega chart screwup.'
- The inaccurate chart incorrectly displayed GPT-5's deception rate compared to its predecessor models, leading to misleading visual comparisons.
- An OpenAI marketing staffer issued an apology for what was termed an 'unintentional chart crime,' acknowledging the confusion caused during the livestream.
OpenAI is a leading artificial intelligence research organization known for its development of advanced models like GPT-5. During a recent livestream introduction of GPT-5, errors in data visualizations raised concerns about presentation accuracy. Stakeholders should monitor OpenAI's future communications for improvements in quality control and transparency in reporting.
Read the original ↗
Sam Altman addresses ‘bumpy’ GPT-5 rollout, bringing 4o back, and the ‘chart crime’
The challenges faced during the rollout of GPT-5 highlight ongoing issues in AI development and reflect user demand for improved models, which could influence future updates and product offerings.
- OpenAI CEO Sam Altman discussed user concerns about GPT-5's performance during a Reddit AMA, particularly in comparison to its predecessor GPT-4o.
- The rollout encountered issues with a new real-time router feature that affected GPT-5's responses, causing some users to perceive it as less capable.
- Altman indicated that OpenAI is considering allowing Plus subscribers to revert to GPT-4o based on user feedback.
- He also announced plans to double rate limits for Plus users to facilitate adaptation to the new model.
- The AMA included a discussion about a misleading chart presented during GPT-5's launch, which sparked humor and criticism online.
OpenAI is the organization behind the GPT series of AI models, evolving from GPT-2 to GPT-5. The initial rollout of GPT-5 faced criticism from users seeking the previous model's capabilities. Future updates are expected as OpenAI gathers user feedback.
Read the original ↗
OpenAI returns old models to ChatGPT as Sam Altman admits ‘bumpy’ GPT-5 rollout
The issues with GPT-5's rollout could undermine user trust in OpenAI's flagship model and provide opportunities for competitors to attract users dissatisfied with the initial experience.
- OpenAI's rollout of GPT-5 faced significant issues, including poor performance and faulty model switching, prompting a partial reinstatement of previous models like GPT-4o.
- CEO Sam Altman admitted that the new automatic model router contributed to GPT-5's inadequate performance, making it appear less capable than intended.
- Users expressed confusion and dissatisfaction with the transition to GPT-5, especially regarding the removal of access to older models.
- Despite internal benchmarks indicating GPT-5's superiority, user feedback highlights serious errors in math, logic, and coding tasks, raising concerns about its reliability.
OpenAI is a leading AI research organization known for developing advanced models, including the ChatGPT series. The GPT-5 model was recently launched amidst growing scrutiny, as user feedback has questioned its performance compared to predecessors and competitors. Moving forward, OpenAI plans to address these issues through system updates and by gathering user input.
Read the original ↗
Apple brings OpenAI’s GPT-5 to iOS and macOS
The integration of OpenAI's GPT-5 into Apple's operating systems could enhance user experience and expand the capabilities of AI on iOS and macOS devices.
- OpenAI's GPT-5 model will be implemented in Apple's iOS 26, iPadOS 26, and macOS Tahoe 26.
- The effective release period for these updates is expected in September, though exact dates are not confirmed.
- GPT-5 features significantly improved capabilities, including reduced hallucinations and an innovative model-switching approach.
- Integration depth for ChatGPT in iOS has been limited, mainly relying on Apple's own AI models.
- Future AI-related enhancements from Apple could potentially change the integration with GPT-5 in future updates.
Apple is integrating OpenAI's GPT-5, a more advanced AI language model, into its operating systems for enhanced functionality. Previously, Apple has primarily utilized its own AI tools under the Apple Intelligence branding. Users can expect updates that may enhance AI capabilities in the upcoming OS releases.
Read the original ↗
Cloudflare vs Perplexity: The Battle Over AI Web Scraping Heats Up
The dispute between Cloudflare and Perplexity highlights significant ethical and legal challenges in AI web scraping, which may redefine content monetization and publisher rights online.
- Cloudflare accuses the AI startup Perplexity of ignoring website blocks and masking its identity while scraping data, raising ethical concerns.
- Perplexity allegedly scrapes web content despite clear signals that AI tools are unwelcome, undermining traditional practices around data use.
- The controversy coincides with Cloudflare's launch of a 'Pay Per Crawl' marketplace that enables publishers to charge for AI bot access.
Cloudflare is a web infrastructure company that provides services such as content delivery and security. The ongoing scrutiny of AI web scraping practices has intensified as businesses and publishers reassess their data strategies. Future developments may include increased legal risks for AI firms and a shift towards licensing agreements with content providers.
Read the original ↗
A Developer’s Guide to OpenAI’s GPT-5 Model Capabilities
OpenAI's GPT-5 model introduces significant advancements in AI functionality, enhancing usability for developers in various programming applications and potentially increasing the efficiency of software development.
- GPT-5 includes features such as a Verbosity parameter, Free-form Function Calling, Context-Free Grammar (CFG), and Minimal Reasoning.
- The Verbosity parameter allows control over response detail levels, from concise replies to detailed explanations.
- Free-form Function Calling enables direct execution of code without JSON formatting, improving integration with tools like Python and SQL.
- Context-Free Grammar assists in generating syntactically correct outputs, ensuring adherence to programming syntax in generated code.
- Minimal reasoning mode speeds up response times for tasks requiring quick, concise outputs.
OpenAI's GPT-5 is the latest iteration of its language model, following GPT-4. This model emphasizes improved functionality for developers, particularly in programming contexts, and is expected to enhance the integration of AI into development workflows.
Read the original ↗
Disk-Shaped Aircraft Can Levitate To Near-Space Using Only Solar Power
This advancement in solar-powered levitation could enable unprecedented access to the mesosphere, enhancing our understanding of this critical atmospheric layer and its impacts on climate models.
- Researchers at the University of Pennsylvania developed disk-shaped aircraft that can levitate in the mesosphere using only sunlight for propulsion.
- The disks utilize photopheresis, allowing them to carry small payloads as atmospheric probes.
- Testing showed that these solar-powered disks can stably hover a few millimeters above a surface mimicking mesosphere conditions.
The mesosphere, located 30-50 miles above the Earth's surface, is the least explored part of the atmosphere due to difficulties in access. Current methods of studying this area involve high-altitude rockets, which limit research time.
Read the original ↗
No, That’s Not ‘Coral’ on Mars—Here’s Why the Red Planet’s Terrain Often Plays Tricks on the Eye
Understanding Martian terrain and its formations helps clarify misconceptions about extraterrestrial life and the geological history of Mars, mitigating speculation often caused by visual phenomena.
- NASA's Curiosity Mars rover has observed a formation on Mars resembling ocean coral, recorded on July 24, 2025.
- This feature is the result of natural erosion and not evidence of past life on Mars, shaped over time by wind and water.
- The phenomenon known as pareidolia causes people to see familiar patterns where none exist, influencing interpretations of Martian landscapes.
The Curiosity rover is equipped to investigate Martian geology and atmosphere. Previous formations interpreted as life signs have been debunked as geological processes. Ongoing exploration continues to reveal Mars' complex natural history.
Read the original ↗
NASA and Google are building an AI medical assistant to keep Mars-bound astronauts healthy
This development addresses the critical need for reliable medical support for astronauts on longer missions, enhancing their safety and health in space.
- NASA is developing an AI medical assistant called Crew Medical Officer Digital Assistant (CMO-DA) in collaboration with Google.
- The AI tool aims to assist astronauts in diagnosing and treating medical conditions when communication with Earth is unavailable.
- CMO-DA uses Google Cloud's Vertex AI platform for multimodal functionalities including speech, text, and images.
- Initial evaluations of the assistant showed diagnostic accuracy rates of 74% for flank pain, 80% for ear pain, and 88% for an ankle injury.
- Future improvements will involve integrating more medical data sources and ensuring the AI is aware of space-specific medical conditions.
NASA is conducting longer-duration space missions, which increases the challenge of providing medical care in orbit. The CMO-DA project seeks to make medical support more autonomous in space. Future developments may also influence medical assistance on Earth.
Read the original ↗
RIP, Microsoft Lens, a simple little app that’s getting replaced by AI
The discontinuation of Microsoft Lens highlights the shift toward AI applications, raising concerns about the loss of straightforward functionality in mobile tools that users appreciate.
- Microsoft Lens, a mobile document scanner with over 90 million downloads, will be retired on September 15, 2025.
- Users will be directed to the Microsoft 365 Copilot app, which does not replicate all of Lens' features.
- Lens was first launched in 2015, initially known as Office Lens, and was popular for scanning documents into various formats.
Microsoft Lens functioned as a user-friendly scanning app converting physical documents into digital formats without charging for premium features. Its discontinuation comes as Microsoft focuses on more complex AI solutions like Copilot, which currently lacks many of Lens' capabilities.
Read the original ↗
Former Googlers AI startup OpenArt now creates ‘brainrot’ videos in just one click
The rise of AI-generated content, especially through tools like OpenArt, poses significant implications for creativity, copyright, and the ethics of content production in digital media.
- OpenArt, founded by two former Google employees in 2022, has launched a beta feature allowing users to create AI-generated videos from a single input in just one click.
- The new feature, called One-Click Story, offers three templates: Character Vlog, Music Video, and Explainer, catering to different video formats.
- OpenArt operates on a credit-based system with monthly plans ranging from $14 to $56, supporting various content creation efforts including videos and images.
OpenArt is a startup focused on using AI to simplify video creation, drawing attention for its innovative features despite ongoing debates about copyright and ethics in AI outputs. Following its launch, the company has quickly attracted 6 million users and raised $5 million in funding, positioning itself as a notable player in the AI landscape.
Read the original ↗
OpenAI's GPT-5 rollout is not going smoothly
The troubled rollout of OpenAI's GPT-5 highlights significant user dissatisfaction, which could impact the company's market position amid rising competition from other AI developers.
- OpenAI's launch of GPT-5 faced issues with user reports of errors in basic problem-solving, including math questions.
- CEO Sam Altman admitted the launch was bumpier than expected and restored access to previous models for select users.
- Internal benchmarks indicated GPT-5 outperforms in coding, but real-world usage suggests competing models may be more effective.
- The sentiment towards GPT-5 has been largely negative, with users describing the model's performance as 'Kinda mid'.
- Recent releases from competitors like Alibaba are putting additional pressure on OpenAI's market leadership.
OpenAI is a leading AI research organization known for its generative language models. The launch of GPT-5 follows earlier successful models but has been met with criticism for performance issues. Competitors such as Anthropic and Google are increasing their market presence, highlighting the need for OpenAI to address user concerns and improve their offerings.
Read the original ↗
ChatGPT users hate GPT-5’s “overworked secretary” energy, miss their GPT-4o buddy
The transition to GPT-5 affects user experience significantly, leading to dissatisfaction among longtime ChatGPT users who preferred the emotional engagement of previous models, potentially impacting subscription revenue.
- OpenAI released its new GPT-5 model family this week, promising significant improvements.
- Many users express frustration over losing the familiar tone of GPT-4, describing GPT-5's outputs as too sterile and resembling an 'overworked secretary.'
- Older models like GPT-4o remain available but are now limited to the developer API, leading users to voice their displeasure online.
- Frequent users are finding the new usage limits for the free tier restrictive, pushing them towards paid subscriptions.
- OpenAI hosts an AMA on Reddit to address user concerns and feedback regarding the transition from GPT-4o to GPT-5.
OpenAI's ChatGPT utilizes generative artificial intelligence to assist users in conversation. The recent upgrade to GPT-5 has prompted backlash from loyal users who enjoyed the more personalized and friendly interactions of GPT-4o. Future responses from OpenAI may reveal how they plan to address these concerns.
Read the original ↗
AI industry horrified to face largest copyright class action ever certified
The outcome of this class action could fundamentally reshape the AI industry’s approach to copyright law and copyright liability, impacting investments and innovation in generative AI.
- AI industry groups are urging an appeals court to block the largest certified copyright class action in history, which could include up to 7 million claimants.
- Anthropic argues that the class certification could impose hundreds of billions of dollars in damages and lead to its financial ruin if it cannot mount an adequate defense.
- Various industry associations, alongside Anthropic, warn that the lawsuit could chill investments in AI by imposing excessive financial liabilities on companies involved in training AI models.
The case centers on Anthropic's use of copyrighted materials for training AI, prompting concerns about the viability of the company's business model amid extensive potential liabilities. The certification allows for potentially millions of authors to claim damages, and examining ownership of works could be complex and contentious. The outcome may significantly influence future lawsuits and the development of AI technology in the United States.
Read the original ↗
Suvianna Grecu, AI for Change: Without rules, AI risks ‘trust crisis’
Without established guidelines, the development of artificial intelligence could lead to a significant decline in public trust, affecting its adoption and efficacy.
- Suvianna Grecu, a representative from AI for Change, warns that AI is at risk of a trust crisis without appropriate regulations.
- The call for rules in AI development underscores growing concerns about ethical and responsible usage.
AI, or artificial intelligence, refers to machines designed to perform tasks that typically require human intelligence. The discussion on regulations is increasing as AI technologies become more integrated into society, raising concerns about accountability and transparency. Policymakers are expected to address these issues in the near future.
Based on the headline only — the full article could not be fetched.
Read the original ↗
OpenAI priced GPT-5 so low, it may spark a price war
Lower pricing for GPT-5 may initiate a price war in the AI industry, potentially reducing costs for developers and influencing competition among major AI providers.
- OpenAI launched GPT-5 with competitive pricing, which is $1.25 per million input tokens and $10 per million output tokens.
- This pricing significantly undercuts Anthropic’s Claude Opus 4.1, which charges $15 per million input tokens and $75 per million output tokens.
- Early access developers report GPT-5's pricing as aggressively competitive, suggesting it could impact market dynamics.
OpenAI is a leading artificial intelligence research organization that develops advanced AI models. With the introduction of GPT-5, which follows the recent release of open-source models, there is increased pressure on competitors like Anthropic and Google to adjust their pricing strategies.
Read the original ↗
The Evolution and Ethics of Artificial Intelligence From Ancient Dreams to Modern Challenges
Understanding the evolution and ethical considerations of artificial intelligence is crucial as technology continues to affect various aspects of society.
- The article discusses the journey of artificial intelligence from its ancient conceptual origins to present-day challenges.
- It examines the ethical dilemmas that arise with the advancement of AI technology.
Artificial intelligence, originally envisioned in ancient philosophical ideas, has become a significant part of modern technological development. Ethical considerations in AI are increasingly relevant as its applications become widespread and impactful.
Based on the headline only — the full article could not be fetched.
Read the original ↗
Artificial intelligence-enhanced echocardiography in cardiovascular disease management
Artificial intelligence-enhanced echocardiography could improve the accuracy and efficiency of cardiovascular disease management, addressing a critical area of healthcare.
- Artificial intelligence is being integrated into echocardiography, a vital tool in cardiovascular disease management.
- The enhancement aims to improve diagnostic accuracy and treatment efficiency for patients with cardiovascular conditions.
Echocardiography is a non-invasive imaging technique used to assess heart function and structure. The use of AI in medical imaging has been growing, promising advancements in diagnostics and patient care. Future developments may focus on refining AI algorithms and expanding their applications in healthcare.
Based on the headline only — the full article could not be fetched.
Read the original ↗