topic
Good news about Artificial intelligence
Positive stories about Artificial intelligence, gathered from the newsrooms Rally reads. Every headline links back to the publisher who reported it.
46 stories.
Colossal Biosciences announces new AI platform to help accelerate de-extinction efforts
Colossal Biosciences has launched an AI platform to speed up its work reviving extinct species. A new fund will also support startups using the technology. The advances could bring lost species back to life.
TechCrunch
Exclusive: Suki, UpToDate Bring AI Clinical Decision Support Into Workflows
Suki's new AI tool surfaces patient-specific clinical guidance, combining medical records with evidence-based info from UpToDate.
Newsweek
Can Safeworld convince people that GenAI robots won’t hurt them?
Safeworld is creating digital humans to ensure robots interact safely with people, addressing concerns about potential harm. The effort could build trust in AI by demonstrating its potential to protect rather than threaten. This work highlights how technology can be designed with human safety in mind.
TechCrunch
Google’s Project Suncatcher AI data center test has officially launched to space onboard SpaceX rocket
Google’s experimental satellite carrying AI chips has launched into space, marking the first step toward orbiting data centers. This could help reduce energy use on Earth by moving computing tasks to space. It’s a promising start for sustainable tech.
Scientific American
Rakovina Therapeutics Partners with Boston-based Celvion Therapeutics to Leverage AI and Real-World Clinical Data for Precision Cancer Drug Development
Rakovina Therapeutics is teaming up with Boston-based Celvion to use AI and real-world clinical data to refine cancer treatments. The collaboration aims to find the best drug combinations, biomarkers, and patient groups for Rakovina’s ATR inhibitor program, potentially speeding up precision medicine. This partnership could improve outcomes for cancer patients by tailoring therapies more effectively.
Financial Post
Anthropic Says Its A.I. Discovered a New Enzyme System That Resembles the Revolutionary Gene-Editing Tool CRISPR
Anthropic’s AI Claude has autonomously identified a new enzyme system similar to CRISPR, a breakthrough in biology. While its practical uses are still unknown, the discovery highlights AI’s growing ability to advance scientific research. This marks a significant step in understanding how AI can contribute to biological discoveries.
Smithsonian Magazine
AI must serve humans, rather than 'become an instrument of domination and injustice', pope warns
Pope Leo XIV urged that artificial intelligence must serve people, not become a tool for control or harm. His call at UNESCO highlights efforts to ensure technology aligns with human values. This push for ethical AI development offers hope for a more just and equitable future.
Euronews
Ukrainian combat units begin testing SPECTR battlefield system
Ukraine’s Defense AI Center A1 is developing an artificial intelligence platform called SPECTR to analyze battlefield data, support operational planning and predict possible Russian actions, the Defense Ministry said on Sept. 25.
NV
Australia’s prime minister aims to rally middle powers and rein in A.I.
Australia’s prime minister is working to unite middle powers to address the risks of artificial intelligence, aiming to create a safer global framework. This collaboration could help countries balance innovation with responsible regulation, benefiting people worldwide. The effort highlights proactive diplomacy to tackle emerging challenges.
The New York Times
UK to fight Russian disinformation and push new global AI standards, Burnham says
The UK is taking proactive steps to combat disinformation and enhance global safety standards for artificial intelligence. By establishing a new center, it aims to foster a more truthful information environment, which will benefit communities by reducing the harmful impact of misinformation. This initiative reflects a commitment to leading international cooperation on critical global challenges.
BBC News
A New Chatbot Wants to Unlock the Secrets in Tattered Ancient Greek Records
Researchers have created an AI tool to help scholars reconstruct damaged ancient Greek papyrus fragments, speeding up a painstaking process. This could reveal new insights into ancient life, preserving knowledge that might otherwise be lost. The technology offers hope for uncovering hidden details in centuries-old texts.
Wired
OpenAI forms math advisory group as its AI resolves more than 100 open problems
OpenAI has created an independent advisory group to give mathematicians a stronger voice in its research. The group, hosted at the Institute for Advanced Study, will help guide OpenAI’s work on math problems without slowing it down. This collaboration could lead to faster progress on open mathematical challenges.
TechCrunch
Why AI Isn’t Likely to Wipe Out Humanity With Bioweapons
Scientists say the risk of AI causing a bioweapons apocalypse is low, easing fears about one of its most extreme threats. This reassurance helps people focus on more pressing concerns while still addressing AI’s potential dangers. It’s a reminder that not every doomsday scenario is likely to come true.
Wired
Could Artificial Intelligence Help With Early Diagnosis of Schizophrenia?
A new AI tool may detect subtle speech differences in people with schizophrenia more accurately than psychiatrists, helping to reduce delays in diagnosis. This could lead to earlier treatment and better outcomes for patients. The technology focuses on how speech changes, offering a more precise way to distinguish schizophrenia from other mental illnesses.
Smithsonian Magazine
The Heart of Engineering Meets the Future of Energy: Sigenergy Showcases AI-Powered Utility-Scale Energy Storage with SigenStack DC-Coupled PV Project in Stuttgart
Sigenergy’s AI-powered energy storage system in Stuttgart shows how smart technology can boost renewable energy use and speed up project timelines. By optimizing energy revenues, the project offers a practical step toward cleaner, more efficient power grids. This innovation helps utilities and communities make the most of sustainable energy.
Financial Post
A new Anthropic model seeks to test how AI could impact the U.S. economy
A new interactive model from Anthropic lets people explore how AI might shape the U.S. economy, helping them test their own predictions. By making complex economic impacts easier to understand, it offers a clear way to assess AI’s potential. This tool gives everyone a chance to engage with the future of work and innovation.
NPR
China’s top court posts guidelines on deepfakes, voice cloning, algorithmic price discrimination
China’s top court has issued guidelines to address risks tied to AI, including unauthorized deepfakes and voice cloning. The rules aim to protect people from misuse of these technologies, offering clearer legal safeguards. This step helps build trust in emerging tech while ensuring fairness and accountability.
Hong Kong Free Press
AMD unveils ‘personal super computer’ for new era of computing
AMD has introduced a "personal supercomputer" designed for AI-driven computing, aiming to redefine how people interact with technology. The Threadripper Halo Station, unveiled at a tech event in Berlin, offers powerful capabilities for individuals and professionals. This innovation could make advanced computing more accessible, opening new possibilities for creativity and productivity.
Euronews
The future of real estate practice in the age of artificial intelligence
AI is transforming real estate by automating property valuation, marketing, and management, helping professionals analyze vast data faster and spot trends. This innovation makes the industry more efficient and accessible, offering new opportunities for investors and homeowners. The shift signals a smarter, more streamlined future for real estate.
Business Day
These Russian Mathematicians Taught AI Models How to Talk to Each Other Without Using Words
Russian mathematicians have developed a way for AI models to communicate without words, using a method they liken to "machine telepathy." The breakthrough could help AI systems work together more efficiently, opening new possibilities for collaboration. It’s a creative step forward in how machines interact.
Wired
'It can outthink me': How a major manufacturer came to embrace AI
GE Appliances is using AI to boost quality and efficiency in its factories, helping workers and the company alike. By embracing smarter technology, it’s making production faster and more reliable, a win for both employees and customers.
NPR
An Anthropic researcher just gave us a peek at self-improving AI
An Anthropic researcher demonstrated an AI system that improved on 10 specific benchmarks without harming overall performance, showing progress in training AI with AI. This could help developers build safer, more reliable models. It’s a step toward more self-improving systems, which could benefit many fields.
TechCrunch
Claude Cowork finally remembers what you told the app in chat
Anthropic has updated Claude’s memory system so it can recall past conversations across chat and Cowork, saving users from repeatedly explaining the same details. This change helps people work more efficiently with the AI, making it a useful improvement for frequent users.
TechCrunch
I spent a day at a robot “carnival” in Shanghai. Here’s what I saw.
China is advancing AI by embedding it into humanoid robots, a priority in its latest five-year plan. The country leads the world in producing these machines, bringing AI closer to everyday life. This progress could make advanced technology more accessible and integrated into daily routines.
MIT Technology Review
Beyond China's humanoid robots, a quieter machine revolution is unfolding
China’s factories are rapidly adopting automation, with over two million robots now at work. This shift is helping workers like a young teacher stay ahead by learning the latest technology to share with his students. The quiet revolution is making advanced skills more accessible, preparing the next generation for an evolving workforce.
BBC News
OpenAI institutes new safeguards after Hugging Face breach
OpenAI is strengthening its security measures to better protect AI models during development and testing. The new safeguards include closer monitoring and a stronger focus on alignment and security, helping prevent future breaches. This move improves safety for users and researchers who rely on OpenAI’s technology.
TechCrunch
DTI, Converge launch AI, digital transformation programs for MSMEs
The Department of Trade and Industry and Converge ICT Solutions have launched programs to help micro, small, and medium enterprises adopt AI and digital tools, boosting their competitiveness. The initiative includes a new center with co-working spaces, training rooms, and tech showcases to support business growth. This effort aims to modernize Philippine businesses and create more opportunities.
Rappler
How kids feel about AI, in their own words
Kids shared their unfiltered thoughts on AI, revealing a mix of creativity and honesty about its role in their lives. Their voices offer a fresh perspective on how young people view technology’s future. The insights help adults understand the next generation’s relationship with AI.
MIT Technology Review
Can AI make mathematics more human?
AI is reshaping mathematical work by making it more accessible and collaborative. Mathematicians like Yang-Hui He see this as a positive shift, as it helps researchers tackle complex problems in new ways. The change could make math more human by broadening its reach and impact.
Scientific American
NUS makes AI courses compulsory, gives all students and staff free ChatGPT Edu access
The National University of Singapore will require all undergraduates to take at least two AI courses before graduating, ensuring students across disciplines gain essential skills. The university is also providing free access to ChatGPT Edu for all students and staff, supporting learning and innovation. This move prepares graduates for a future shaped by AI.
The Straits Times
AI for science needs reasoning, not just data
AI is advancing beyond just processing data—it’s now learning to reason, which could unlock new scientific breakthroughs. This shift moves beyond past claims that science is nearing its limits, offering hope for deeper discoveries. The progress could help researchers tackle complex problems in ways that pure data analysis couldn’t.
MIT Technology Review
DeepMind’s hurricane breakthrough has surprised weather scientists
AI weather models like WeatherNext can now make accurate predictions using lower-resolution data, helping scientists forecast storms with greater confidence. This breakthrough could improve early warnings and preparedness, especially in vulnerable regions. The advance shows how AI is making weather science more precise and accessible.
Ars Technica
Why Google DeepMind broke up the AlphaFold team
The artificial intelligence system largely accomplished the task it was built to solve, and researchers outside DeepMind have already been carrying its work forward
Scientific American
A.I. Went to the Opera, Creating Images for an Anniversary Production of Wagner's 'Ring Cycle' at the Theater Where He First Staged the Masterpiece
An experimental production of Wagner’s "Ring Cycle" at the Bayreuth Festival uses AI to reinterpret the classic opera, offering a fresh take on its artwork. The project honors the opera’s legacy while embracing modern technology, giving audiences a new way to experience a timeless work.
Smithsonian Magazine
Why winner of biggest prize in maths has decided to work on AI instead
A Fields medal-winning mathematician is shifting his focus to AI safety, recognizing the technology’s potential to transform lives—and its risks. His expertise could help ensure AI develops in ways that benefit everyone. The move highlights how urgent and impactful this work has become.
New Scientist
Yang Zhilin, the rock enthusiast dreaming of AI’s other side
A 34-year-old founder has unveiled an AI model that narrows the gap between China and the U.S. in the technology race. The breakthrough could advance global innovation by making cutting-edge AI more accessible. It’s a promising step toward a more competitive and collaborative field.
El País (English)
Microsoft launches its first cybersecurity model, plus a new agentic cybersecurity system
Microsoft has introduced its first AI security model and a new cybersecurity platform, aiming to strengthen digital defenses. The move targets major competitors and could help businesses better protect their systems. This is a step forward in making online security more robust and accessible.
TechCrunch
U.S., other nations back open-source AI with 'strong security' at China summit
Governments at the APEC summit agreed to collaborate on open-source AI with strong security, marking the first time such cooperation was included in a minister-level statement. This step could help ensure AI advancements are both accessible and safe, benefiting developers and users worldwide. The move signals growing international support for responsible AI development.
CNBC
AegisAI, founded by former Google security execs, lands $36M to stop AI-driven spear phishing
AegisAI’s AI agents analyze emails for subtle anomalies that automated checks might miss, helping businesses defend against AI-powered spear phishing. The startup’s $36 million funding will expand protection as hackers increasingly use AI to craft convincing, personalized attacks. This could make online communication safer for everyone.
TechCrunch
'Stopping AI because it carries risks? That would be like banning cars because they cause accidents'
Instead of halting AI over risks, experts argue for building public trust to guide its development. This approach could help balance innovation with safety, ensuring technology serves society responsibly. The focus shifts from fear to constructive progress.
Le Monde
This Former Intel CEO Wants to Jumpstart Moore’s Law With Light
Pat Gelsinger is exploring ways to advance computing power using light, which could support more advanced AI. His approach aims to push beyond current limits, offering a promising path for future innovation. The work could help industries relying on cutting-edge technology.
Wired
Man of his word: Pope Leo speeches declared human-authored by Australian AI detection tool
Maps of Hope has been certified by Proudly Human, a company led by former chief scientist, Dr Alan Finkel
The Guardian
Beyond grep: The case for a context-rich AI coding harness
AI-powered coding tools are getting smarter, but the real innovation lies in the software that manages them. Developers now have better harnesses to work with these models, making coding more efficient and context-aware. This shift could streamline how programmers build and debug, saving time and improving accuracy.
Ars Technica
Rome declaration for an ‘unarmed and disarming peace’ signed by Nobel laureates, experts
The Global Nobel Laureates Assembly on Artificial Intelligence and Nuclear War culminates in the drafting and signing of the Rome Declaration for an Unarmed and Disarming Peace
Rappler
Meet GPT-Red: an LLM super-hacker OpenAI built to make its models safer
OpenAI has developed GPT-Red, an LLM designed to test and strengthen its models against cyberattacks. The latest version, GPT-5.6, was trained using this adversarial approach to improve safety. This proactive step helps protect users from emerging threats.
MIT Technology Review
Thinking Machines Lab Drops Its First Model
Inkling, a 975-billion-parameter open source model, was trained to understand video and audio. It could help Thinking Machines establish itself among competitors like Anthropic and OpenAI.
Wired
