Top 10 AI News
Latest AI and Technology News
Legal AI Startup Legora Is In Talks To Raise Funding At A $1.8 Billion Valuation
Just four months after its Series B round, Legora, a Stockholm-based legal AI startup, is picking up more capital. The company, which uses AI to streamline legal processes, is in talks to raise funding that would value it around $1.8 billion.
Meet The Stanford Dropout Building An AI To Solve Math’s Hardest Problems—And Create Harder Ones
Axiom Math, founded by a Stanford dropout and staffed with former Meta engineers, has raised $64 million in seed funding. The startup is building an AI “math whiz” designed to solve extremely challenging math problems and even generate new ones.
Lufthansa Bets Big On AI To Cut 4,000 Jobs
Lufthansa has announced plans to cut roughly 4,000 administrative jobs by 2030. The airline will replace many traditional office roles with digital processes and AI-driven applications, betting on automation to improve efficiency and future profits.
The Good And Bad News For Brands About Google’s New Agentic AI Browser
Google Chrome has been reimagined with integrated AI features, marking the start of the “agentic” browsing era. This transformation brings both opportunities and challenges for brands: enhanced personalization and engagement for marketers, but also increased complexity in competing for consumer attention.
Tesla’s Full-Self Driving Software Is A Mess. Should It Be Legal?
Elon Musk has been touting Tesla’s AI-enhanced Full Self-Driving (FSD) system as a key safety feature and sales driver. However, a recent public test revealed serious flaws: the FSD software ignored road signs and even ran over a mannequin child. This raises new questions about the legal and safety risks of Tesla’s partially automated driving technology.
No Pixel 10 Needed: Google Photos’ Conversational Editing Comes To All Android Devices
Google has extended its AI-powered photo editing features to all Android users, not just Pixel 10 owners. Users can now use conversational commands in Google Photos to enhance and modify images—essentially chatting with the app to make edits like adjusting lighting or removing objects.
LLMs Are Short-Circuiting. Is It Time To Redefine Intelligence?
Many in the AI community are sounding alarms: large language models (LLMs) are failing basic tasks in ways that suggest they may not deliver on the promise of true intelligence. As these models “short-circuit” or hallucinate more often, researchers are questioning whether we need a new definition of machine intelligence—and a new approach beyond just scaling up models.
Frontier Careers: Where Will New AI Jobs Be?
Studies suggest AI will replace or impact up to 50% of white-collar jobs, but it will also create new roles. This article explores which industries and fields are likely to see “frontier” AI jobs emerge—areas like AI ethics, data curation, hybrid human-AI roles, and specialties in AI-driven healthcare or law.
Andrew Ng’s LandingAI Develops Specialized Model To Ease Document Intelligence
AI pioneer Andrew Ng’s startup, LandingAI, is targeting enterprise document processing. Its new platform, ADE DPT-2, aims to make document understanding more agentic—organizing, analyzing, and acting on business documents automatically. The goal is to use the AI as the “backbone” of document workflows, potentially transforming industries that handle large volumes of paperwork.
Banking On Trust: FICO Launches Financial Foundation Models
Credit scoring firm FICO announced new AI models specialized for financial services. They claim their narrow, finance-focused foundation models will outperform generic AI in banking and lending tasks. Their launch includes a “Focused Language Model” for finance, designed to handle complex, sector-specific data like credit histories and risk profiles.
Synthetic Personas Offer The AI Edge Your Sales Team Needs
AI-generated “synthetic personas” are being marketed as the salesforce of the future. These are AI-crafted customer avatars with detailed behaviors and preferences. The idea is that sales teams can use them to predict customer responses and tailor pitches, effectively running dynamic simulations of real-world buyer behavior.
ServiceNow Says AI-First Interfaces Will Define The Future Of Enterprise AI
Enterprise software maker ServiceNow unveiled its vision for an “AI Experience” platform. The company argues that AI-first interfaces—where users interact with tools primarily through natural language and intelligent assistants—will become the new standard. ServiceNow is positioning itself to compete with Microsoft and Salesforce for dominance in the emerging agentic AI-driven enterprise market.
Workday Rising 2025 Showcases Workday’s Pivot To AI Platform
At its annual user conference, Workday announced a strategic shift toward AI. The company plans to build a domain-specific AI platform (e.g., for HR and finance) and leverage an open data ecosystem so partners can integrate AI tools. The goal is to evolve Workday into an enterprise AI platform managing people, money, and automated agents.
Precision Mental Health Gets Precisely Boosted Via Innovative Uses Of Advanced AI And LLMs
Stanford researchers recently presented work on “precision mental health,” using AI to tailor mental health interventions. This article highlights their findings and discusses how advanced AI and language models can personalize care—an exclusive scoop from an AI research consortium symposium.
The Rise Of AI-Powered Ransomware: The Need To Adapt
Cybersecurity experts warn that ransomware is evolving. No longer just a hacker’s tool, next-gen ransomware uses AI to evade detection and automate attacks, effectively turning cybercrime into a sophisticated weapon. Businesses and governments are being urged to adapt their defenses against this new, AI-enhanced threat.
Are Uber And Lyft Becoming Obsolete?
This “real-world scenario” article explores how emerging AI-driven transportation could disrupt ride-hailing. It uses current trends to predict the future: autonomous vehicles, drone taxis, and AI-powered traffic management might soon render traditional Uber and Lyft models outdated.
New Snapdragon Mobile And PC Platforms Set To Fuel A New Breed Of Agentic AI
Qualcomm announced its upcoming Snapdragon chips for mobile and PC, designed for “agentic AI.” These chips will not only execute user commands but also run AI models locally. Qualcomm’s vision is a future where devices proactively understand context and assist users without relying solely on cloud computing.
Responding to the Climate Impact of Generative AI
The explosive growth of AI data centers worldwide is projected to increase greenhouse gas emissions significantly. MIT researchers are now investigating ways to reduce the environmental harms of training giant models, seeking energy-efficient algorithms and sustainable hardware solutions to mitigate AI’s carbon footprint.
AI System Learns from Many Types of Scientific Information and Runs Experiments to Discover New Materials
MIT scientists unveiled “CRESt,” an AI platform that can assimilate diverse scientific data – from research papers to lab notes – and autonomously run lab experiments. CRESt was used to discover novel materials for energy applications, potentially speeding up solutions to long-standing engineering challenges.
New AI System Could Accelerate Clinical Research
An innovative AI tool developed by MIT can rapidly annotate medical images at scale. This means researchers can quickly identify tumors or other disease markers in scans, accelerating the pace of clinical trials and medical research without having to manually label millions of images.
Improving the Workplace of the Future
Economics PhD student Whitney Zhang of MIT is studying how new technologies like AI, and company policies around them, shape the job market. Her work looks at how automation and organizational decisions impact jobs, aiming to ensure better outcomes as workplaces evolve with AI.
MIT Affiliates Win AI for Math Grants to Accelerate Mathematical Discovery
Two MIT mathematicians, David Roe and Andrew Sutherland, received grants to apply AI in math research. Their goal is to advance automated theorem proving—letting AI help prove complex mathematical theorems and discover new insights, a project carried out in collaboration with MIT alum and other researchers.
New Tool Makes Generative AI Models More Likely to Create Breakthrough Materials
Researchers introduced SCIGEN, a new technique that steers generative AI models toward designing materials with desired exotic properties. For example, SCIGEN can guide an AI to propose materials for quantum computing or renewable energy that it might not otherwise find, opening new frontiers in materials science.
How Are MIT Entrepreneurs Using AI?
At MIT’s delta v summer accelerator, startups showcased how they are embedding AI in their businesses. The feature highlights several young companies using AI for everything from logistics optimization to personalized health apps, illustrating how the next generation of entrepreneurs is leveraging AI innovation to solve real problems.
How to Build AI Scaling Laws for Efficient LLM Training and Budget Maximization
Scientists from MIT and IBM Watson have devised a guide of universal “scaling laws” for AI. These formulas allow engineers to predict how larger language models will perform based on smaller ones, helping teams plan their compute budgets and tune performance without needless experimentation.
OpenAI’s Sora joins Meta in pushing AI-generated videos. Some are worried about a flood of ‘AI slop’
OpenAI has launched Sora, a new AI-driven video app aiming to compete with TikTok and Instagram. Users can quickly generate short videos with AI. While exciting for content creators, critics fear it could lead to a deluge of low-quality or misleading AI-generated videos, saturating social media feeds with what some call “AI slop.”
OpenAI’s ChatGPT now lets users buy from Etsy, Shopify in push for chatbot shopping
OpenAI added a new shopping plugin to ChatGPT, enabling users to browse and buy products from marketplaces like Etsy and Shopify directly within the chat interface. This move is part of an industry trend to blend e-commerce with conversational AI assistants, allowing for instant search and checkout via chatbot.
Nvidia to invest $100 billion in OpenAI to help expand the ChatGPT maker’s computing power
Chipmaker Nvidia announced a landmark partnership with OpenAI, pledging $100 billion in investment. Nvidia will build at least 10 gigawatts of new AI data center capacity to power the company behind ChatGPT. This massive infrastructure expansion is aimed at meeting the exploding demand for AI computing power.
California Gov. Gavin Newsom signs landmark bill creating AI safety measures
California passed a first-of-its-kind law requiring companies to perform risk assessments on their AI systems. Gov. Newsom signed legislation that mandates AI developers to account for bias and safety in new systems. The bill positions California as a national leader in AI regulation, aiming to protect citizens from unchecked AI risks.
AI’s double-edged sword: UN leaders weigh its promise and peril
At a high-profile summit in New York, United Nations leaders and AI experts debated the revolution of artificial intelligence. The discussions highlighted AI’s incredible benefits in fields like medicine and climate, but also warned of dangers: deepfakes, job displacement, and autonomous weapons. Consensus: global cooperation is needed to navigate AI responsibly.
Meta unveils AI-powered smart glasses with display and neural wristband at Connect event
Meta CEO Mark Zuckerberg revealed a new line of “smart glasses” at the Meta Connect conference. These glasses project information in the wearer’s field of view, powered by on-device AI. A companion “neural wristband” uses muscle signals for control. The devices represent a push toward integrating AI seamlessly into everyday wearables.
One Tech Tip: OpenAI adds parental controls to ChatGPT for teen safety
OpenAI announced new safety features for minors using ChatGPT. The updates include a “Teen mode” that filters profanity and inappropriate topics, as well as usage limits tied to a gamerkid.com-age verification. These parental controls are intended to make chatbot use safer for younger users.