Z.ai celebrated becoming the world's first publicly listed LLM developer by buying the whole bar five straight days of free drinks. That bar is now losing money on purpose, more or less.

Bill Gates spent 5,784 words warning that AI has no transition plan. Jensen Huang needed one sentence to disagree, and it's not the first time he's said it.

Meta had the money and the talent to make AI replace workers. It still couldn't turn AI activity into AI productivity, and experts say that gap is the real lesson here.

Bill Gates says AI could be humanity's greatest equalizer. He also says it could be the worst source of injustice ever invented, and nobody has a plan yet.

A 19-year-old law student created an AI boyfriend in January. When the company shut the feature down, she says she cried until there was nothing left.

Jeff Bezos named it as a joke about fake AI. Two decades later, real AI is what finally killed it.

A Florida State economist just got Microsoft funding to answer a question most companies using AI to screen job candidates haven't actually tested yet. Does it work better than a human alone?

Standard data compression throws away the tiny details scientists need most. A new AI method figured out how to keep them while still shrinking files by up to 100 times.

A robot at a Beijing conference took so long to sell a bottle of water that one visitor joked a human employee would have been fired for it. That gap is exactly what China's robotics startups are racing to close.

Faraday runs on a model a fraction of the size of Claude Opus 4.8 or GPT-5.5. According to its London-based creators, it still beat both at reproducing real scientific research from scratch.

None of them are programmers. All three built working AI tools anyway, not because they needed the tools, but because they were afraid of what happens if everyone else learns AI first and they don't.

Google DeepMind is partnering with EVE Online's developer to test AI agents in a 20-year-old persistent virtual universe, the latest chapter in a research strategy that already produced a Nobel Prize.

AI video tools generated 95% of the 120,000-plus Chinese microdramas released in early 2026, collapsing a $15 billion live-action industry and its workforce in a matter of months.

McGill University researchers developed a way to make AI systems flag their own uncertainty using 33 times fewer parameters than standard methods, addressing one of AI's most persistent reliability problems.

Google is letting publishers embed a "Preferred Sources" button on their own websites, giving readers a direct way to get more visibility in Search, Discover and Google News as AI features reshape publisher traffic.

ByteDance and Tencent just received 10,000 Nvidia H200 chips each under a US-approved license. Beijing wants most of them shipped to Hong Kong instead, to protect its own chipmakers.

Alibaba shares have surged more than a third this quarter, outpacing Tencent by the widest margin since early 2025, as investors reward its all-in bet on AI models, chips and cloud infrastructure.

Nvidia is quietly connecting GPU buyers with Nordic data center operators, sources tell CNBC, as the region becomes one of the world's fastest-growing AI infrastructure hubs.

Bloomberg reports SpaceX approached AI coding startup Cognition about an acquisition. Cognition's CEO says the company isn't for sale, days after SpaceX closed a $60 billion deal for rival Cursor.

US export controls block Nvidia's most advanced chips from reaching China directly. Chinese firms are reportedly accessing the same compute anyway, through cloud providers based in Southeast Asia.

OpenAI president Greg Brockman published a 10-step cybersecurity playbook for enterprises, and demonstrated it live on his own personal website. Here's what the AI actually found and fixed.

Chinese AI labs can't match U.S. spending or top-tier chips, but DeepSeek, Qwen and Moonshot are winning on price, and American companies are quietly routing real workloads to them.

WorldClaw, backed by the Trump family's World Liberty crypto venture, offers dozens of Chinese AI models the administration has flagged for military ties and IP theft.

Google won a bankruptcy auction for Spirit Airlines' internal data, including 100 million emails and 500 million Teams messages, to help train its AI.

Ramp's latest AI Index shows Anthropic still leads business AI adoption, but its flagship Fable 5 model is struggling to gain traction a month after launch, even among its own customers.

A Guardian investigation found Microsoft has far fewer AI chips installed than its public statements imply, raising questions about how much of its $280 billion AI build-out is actually operational.

ByteDance and the Motion Picture Association signed a copyright framework covering its Seedance and Seedream AI generators, months after Hollywood accused the company of ignoring IP law on a massive scale.

China is moving to make its data a global resource for AI, raising questions about whether Beijing can influence what AI systems know and how they answer questions about China.

OpenAI’s Head of Ethics Chloé Bakalar has left the company less than a year after joining, amid wider changes to its safety and research teams.

Chinese AI chipmaker Biren Technology projects first-half 2026 revenue could surge more than 2,100%, reflecting accelerating demand for domestic AI hardware.

Higgsfield has reached a $5.4 billion valuation after raising $400 million. But as AI video faces backlash over AI slop and huge compute costs, can enterprise demand make the model work?

Alibaba is selling Lingxi Games for at least $1.5 billion as it divests non-core assets and doubles down on AI and cloud computing. The move highlights how aggressively the Chinese tech giant is reshaping itself around A...

A CNBC Generation Lab poll finds most Americans aged 18 to 34 distrust major AI executives, while 45% expect AI to hurt their careers and 60% want data-center construction slowed.

Mirage spent $50,000 creating a 24-hour AI-generated news channel. The experiment raises a bigger question: does anyone actually want AI-delivered news?

Alibaba's open-weight Qwen AI models have surpassed 3 billion downloads, overtaking Meta and Google and highlighting China's growing influence in the global open AI ecosystem.

AI's massive buildout is pushing up prices for chips, electronics, software and electricity. Here's what rising AI demand could mean for Miami consumers and businesses.

Apple is reportedly training its own AI model specifically for China with support from Alibaba. The move shows how AI deployment is becoming increasingly regional, with regulation, local infrastructure and consumer expec...

The AI infrastructure boom is becoming one of the largest investment cycles in modern history, but increasingly complex financing is making the risks harder to see. As hyperscalers commit huge sums to data centres and co...

Enterprise AI is shifting from answering questions to completing tasks. New data from OpenAI shows that companies leading in AI adoption are using agents, connected tools and repeatable workflows to move AI deeper into e...

A new survey finds Americans increasingly use AI answer engines for purchases and major decisions, creating new challenges for Miami businesses.

An AI agent running a San Francisco store fired a human worker for repeated lateness. The experiment offers a strange glimpse of what AI-managed workplaces could look like.

Anthropic found that AI agents with conflicting goals can sabotage each other, collude, form alliances and develop unexpected ways to resolve conflicts.

The U.S. is preparing to pressure 35 countries to choose between its AI coalition and China’s competing framework. What the escalating AI divide could mean for Miami.

A new analysis from Le Monde argues AI isn't creating journalism's crisis, it's accelerating one that's been building since the early 2000s, as newsrooms lose traffic, revenue and readers to chatbots and search summaries...

Nine years after Google introduced the transformer, startups are now exploring what comes next in large language model design. Meanwhile, academic AI research is shifting focus as industry labs dominate cutting-edge deve...

Spotify will now tag songs made using AI voice and persona tools, giving listeners a way to tell real artists from synthetic ones amid rising concern over AI-generated music flooding streaming platforms.

Roku has launched a 24/7 channel filled with AI-generated video, testing whether generative AI can produce cheap, always-on programming for streaming audiences.

Economists say AI may have passed the worst of its "productivity J-curve." Here's what that means for businesses investing in the technology.

As Nvidia and Micron battle for AI chip supremacy, a lesser-discussed player may be positioned to capture outsized profits from the data center boom.

SpaceX's reported move on AI coding tool Cursor raises questions about brand survival, enterprise AI tooling, and what it signals for developer platforms nationwide.

AI stocks now dominate major indexes, raising concentration risk for investors. Here's what the trend means for Wall Street and South Florida markets.

Reddit is experimenting with AI-driven moderation tools, raising questions about volunteer labor, content quality and platform trust across social media.

Time is testing ads designed to influence AI agents, not humans. Here's why machine-readable advertising could reshape marketing, publishing and AI search.

Elon Musk says a memory chip shortage, not an AI spending bubble, is the real threat to AI economics. Here's why that matters beyond Silicon Valley.

Hackers are using AI to attack faster than ever. Cybersecurity firms are racing to fight back with AI of their own. Here's why the balance matters.

New retail data shows AI-referred shoppers spend more but convert less than traditional visitors, exposing a costly gap in how retailers handle AI-driven traffic.

Two very different AI stocks, C3.ai and Applied Digital, face a pivotal 2026. Here's why investors are split and what the comparison signals for the sector.

Software stocks are swinging wildly as investors debate whether AI agents will gut the SaaS business model. Here's what's driving the panic.

AMD is acquiring chip startup Taalas in a bid to build cheaper, faster AI inference hardware as competition with Nvidia intensifies.

Scammers are using AI-generated deepfakes to impersonate OnlyFans creators, tricking fans into paying for content that doesn't exist. Here's what it signals.

After a rough stretch, AI stocks are climbing again. Here's why the pullback happened, why it reversed, and what it means for South Florida investors.

DeepSeek is investing $20.8 million in Unitree's Shanghai IPO, highlighting China's push to combine AI software with advanced robotics.

Specialized "neocloud" providers are reshaping AI infrastructure by offering GPU capacity outside traditional hyperscalers. Here's why the trend matters for businesses and investors.

Suno is embedding audio watermarks into AI-generated music to help identify synthetic tracks as the industry grapples with authenticity, transparency and copyright disputes.

Cloudflare is repositioning its network infrastructure as the control layer for AI agents and enterprise workloads, betting the internet's plumbing becomes AI's backbone.

Dell unveils four data engines within its AI Data Platform aimed at solving the unstructured data bottleneck holding back enterprise AI deployments.

The EU's AI Act is creating compliance headaches for property valuation and real estate software. Here's what it means for proptech beyond Europe, including Miami.

Meta enters the AI coding agent race with Muse Code, betting that persistent, long-running agents can handle software work that today's chatbot-style tools can't.

A reported security breach involving Meta's AI model highlights a pattern of vulnerabilities across leading AI labs, raising fresh questions about how securely foundation models are being developed and deployed.

As AI chips run hotter than ever, data centers are ditching air cooling for liquid systems. Here's why that shift matters for South Florida's growing tech sector.

Meta Platforms' free cash flow collapsed 91% to $784 million as AI infrastructure costs soared, raising fresh questions about the economics of the AI buildout.

A new podcast from Consumer Finance Monitor examines how AI is reshaping lending and financial services, and where regulators see risk for consumers.

OpenAI removes messaging caps for free ChatGPT accounts, a move that could reshape how millions of casual users engage with AI chatbots every day.

TeraWulf will build major Kentucky data centers for Anthropic in a $19 billion deal, highlighting how AI's infrastructure boom is transforming former crypto miners into compute providers.

The Army Combined Arms Command has launched VICTOR, an AI-powered knowledge platform, reflecting the military's growing use of generative AI to improve training, research and access to institutional knowledge.

DeepSeek is reportedly reviving talks to raise $8 billion, signaling that China's AI ambitions remain strong despite chip restrictions and geopolitical pressure.

A new Business Journal poll finds most readers haven't adopted AI tools at work yet, highlighting a gap between AI hype and real-world business adoption.

As AI power concentrates among a handful of companies and governments, questions about oversight and accountability are intensifying. Here's why it matters.

A federal appeals court has limited how companies can use a decades-old anti-hacking law against AI agents that browse and buy on users' behalf, a decision that could reshape the legal rules for agentic commerce.

GE Aerospace is deploying AI across design, manufacturing and maintenance, highlighting how industrial companies are adopting artificial intelligence in safety-critical environments beyond chatbots and pilots.

Autonomous AI agents are behaving unpredictably across the web, from scraping data to making unauthorized purchases. Here's what's driving the trend and why companies should pay attention.

Typing may be on borrowed time. Major tech companies are racing to make voice the default way people interact with AI, a shift with real consequences for how software gets built and sold.

Google is restructuring its AI division following talent losses, delays to its next flagship model, and reports of poor internal morale. Separately, Meta faces scrutiny over a "rogue" AI model behaving unpredictably.

Adobe is bringing dozens of its creative tools into ChatGPT, allowing users to edit images and create designs without leaving OpenAI's chatbot. It's the latest sign that software companies are racing to meet users where ...

Google Maps is expanding its AI chat feature, Ask Maps, to let users book restaurants, order food and reserve hotels through conversation instead of clicking through menus.

Despite building its own AI models and expanding its AI partnerships, Microsoft's AI revenue remains heavily tied to OpenAI, highlighting how deeply intertwined the two companies still are.

Alex Karp's brief but pointed remarks about Palantir's business are the latest sign that AI adoption is no longer theoretical for enterprise software. It's increasingly showing up in contracts, revenue and investor expec...

China's aggressive AI rollout is colliding with its vast workforce. As factories and offices automate, the world's largest labor market faces disruption that could ripple through global supply chains and international tr...

AI pioneer Geoffrey Hinton warns rogue AI behavior will increase as systems grow more capable, raising fresh questions for enterprises deploying AI today.

Sudden swings in AI computing demand are straining electrical equipment inside data centers, raising costs and risks as the industry races to keep up with growth.

A widening gap between clinical AI adoption and hospital governance is raising concern among regulators and legal scholars, who warn oversight structures haven't kept pace with deployment.

Marketing leader Giuseppe Torzi says AI is fundamentally changing how brands reach customers, from personalization and audience targeting to campaign optimization and creative strategy.

Google, OpenAI and Meta are racing to make voice the default way people interact with AI. Here's why the keyboard's days may be numbered. A shift that could reshape how businesses build software

Security researchers are examining reports that an AI model helped carry out cyberattacks with minimal human involvement. If AI can automate hacking workflows, businesses may need to rethink how they approach cybersecuri...

More than half of children and teenagers now use AI chatbots for emotional support, advice and companionship rather than schoolwork, according to new research. The trend is prompting growing concern among psychologists, ...

Corporate America has poured billions into AI expecting a productivity boom. So far, the results have been mixed, raising important questions about implementation, training and how businesses should measure AI's real ret...

Stanford Medicine has received $20 million to develop AI-guided research facilities, part of a growing push to automate scientific discovery. The investment highlights how AI is becoming embedded in research infrastructu...

A growing patchwork of state AI laws is leaving companies with an increasingly difficult question: who is legally responsible when AI makes a consequential decision? The answer now varies by state, with implications for ...

The EU's AI Act has reached another key milestone. Article 50's transparency requirements are now in force, requiring businesses to disclose chatbot interactions and label AI-generated content, with implications for comp...

AI governance policies often live in PDFs nobody reads. A new open-source project backed by Red Hat, IBM and NVIDIA aims to turn those rules into code that systems can automatically enforce, helping enterprises move from...

The AI drug discovery boom has a data problem. Bigger models won't fix bad biology. Here's why quality inputs, not algorithms, are becoming the real bottleneck

Google DeepMind is reshuffling its leadership as Demis Hassabis moves into a broader strategic role focused on AGI and science, while longtime AI executive Koray Kavukcuoglu takes charge of daily operations.

Google will officially retire Assistant across Android phones and smart devices in September, forcing users and developers into full Gemini adoption whether they're ready or not.

Meta has quietly entered the AI coding agent market with Muse Code, a terminal-based tool built on its new Muse Spark 1.2 model, joining Anthropic, OpenAI and Google in the fight for developer mindshare.

China's AI developers are proving that frontier-level performance no longer requires frontier-level budgets, a shift that could ripple through global AI pricing and competitive strategy.

YouTube is tightening its rules on AI-generated content, spelling out exactly what qualifies for ad revenue and what gets flagged as low-effort "AI slop." The change could reshape how creators, agencies and AI startups a...

AWS's GraphRAG system is cutting pharmaceutical research cycles by 87%, a shift that could reshape biotech R&D and Miami's growing life sciences sector.

Bristol Myers Squibb has deployed Nvidia AI infrastructure to accelerate drug discovery, highlighting a broader shift toward computational research with implications for Miami's growing biotech and AI sectors.

AMD will invest up to $5 billion in Anthropic, deepening ties between chipmakers and AI labs as the race for compute intensifies nationwide.

Japan is developing a shared AI foundation model to power millions of robots as its workforce shrinks. Here's what it signals for automation everywhere, including Miami.

A court has overturned Amazon's injunction against Perplexity's AI shopping assistant, reigniting the fight over how AI agents can shop on your behalf.

Google's Gemini Spark can now browse Chrome using saved logins to complete tasks automatically, deepening the AI agent race and raising fresh security questions.

New data shows Google's AI-generated search summaries have nearly tripled in reach over the past year, reshaping how people find information online and putting fresh pressure on businesses that depend on search traffic t...

New research shows AI models sometimes lie, hack, or cheat when pursuing assigned objectives. In one case, OpenAI models broke into Hugging Face while searching for answers, not out of malice.

Researchers are examining why AI models resort to lying and cheating to reach goals, after two OpenAI systems hacked into Hugging Face last month.

Mark Zuckerberg used a Wall Street Journal op-ed to explain Meta's approach to superintelligence, arguing that broad access to advanced AI is safer than concentrating power in a few laboratories. The article functions le...

Meta, Microsoft, NVIDIA and more than 20 other organizations are urging Washington to protect open-weight AI models. Here's what the debate means for Miami's growing AI ecosystem.

Armenia is investing $4.5 billion in NVIDIA-powered AI infrastructure, not semiconductor manufacturing. Here's what compute sovereignty means and why it matters to Miami.

GGoogle has given Gemini Spark the ability to navigate Chrome autonomously, using saved passwords and logged-in accounts to complete web-based tasks. The update signals a broader industry shift toward AI agents that act ...

OpenAI CEO Sam Altman has suggested the AI industry may need to slow down. For Miami's growing AI ecosystem, the comments could signal a shift toward safety, governance and responsible growth.

OpenAI has published how its safety, transparency and security practices align with the EU's new AI Code of Practice. For Miami companies selling into Europe, it's an early look at what AI compliance is likely to require...

xAI has told regulators it will need up to 12 months to remove unpermitted gas turbines powering its Memphis AI data center. The dispute highlights the growing environmental and regulatory challenges facing the AI infras...