AI News
What's happening in artificial intelligence, explained for everyone
AI agents arrive on your mobile and at work: what they are and why they will change your day to day
Autonomous AI agents are now available on mobile devices and business applications, allowing machines to perform complex tasks without constant human intervention. It is the next evolution after ChatGPT: AI that not only responds, but acts.
Google launches Gemini Omni 1.1 Flash: more speed and lower cost
Google updated its Gemini Flash model with improvements in speed and price reduction, making it more accessible for companies that use AI in daily operations. It is the best option for applications that need quick responses without sacrificing quality.
Google launches Gemini Transcribe: converts audio to text automatically
Google added a new capability to Gemini to automatically transcribe audio files, voice, and videos without needing external tools. This function allows professionals to save hours processing recordings of meetings, interviews, and calls.
The turbulent era of AI is already here: the decisions your company must make now
Experts warn that we are entering a critical phase where AI becomes more capable but also more unpredictable, and companies that don't make strategic decisions now will fall behind. The coming months will determine whether your organization leads or follows in the AI transformation.
Simple Tasks That Surprise AI: What You Should Know Before Using It
Although AI is increasingly powerful, there are simple real-world tasks that completely confuse it. Understanding these limitations is crucial for using AI effectively without frustration.
AI Agents Now Work on Your Phone: What It Means for Your Productivity
Intelligent AI agents, which until now only worked on computers, are now available on smartphones. This means you can have a powerful AI assistant in your pocket that automates tasks without needing constant internet.
OpenAI Presents Jalapeño: More Powerful and Efficient AI Than Before
OpenAI has launched Jalapeño, a new AI model that outperforms previous systems, promising better speed and capabilities without increasing costs. This update means your AI tools will work faster and with better accuracy in daily tasks.
AI now learns to code designs: what it means for designers and developers
Researchers have succeeded in training AI models to write code from visual instructions, combining the creativity of a design with the precision of automatic programming. This is an emerging trend that could significantly accelerate project development speed.
OpenAI lowers prices of GPT 5.6 Sol: save more on your AI projects
OpenAI has just reduced the cost of using GPT 5.6 Sol, its most advanced model, in a promotion that will last until at least November. This means your AI projects, analyses, and automations will be more economical over the coming months.
Claude loses users to cheaper tools: is it worth paying more?
Anthropic's most advanced model (Claude) is struggling to attract users because there are more economical alternatives on the market. This is a key moment to evaluate whether you really need to pay premium or if cheap options are sufficient for your work.
DeepSeek launches ability to 'see' images: how this changes what you can automate
DeepSeek-v4-flash-vision is a new capability that allows AI to analyze images and videos like ChatGPT and Claude do. This opens possibilities to automate tasks that previously required human eyes: reviewing documents, inspecting products, or validating content.
OzBrain: The tool that allows your team and AI to share knowledge
OzBrain is a new platform that functions as a 'shared memory' between your employees and AI agents, allowing everyone to work with the same information. It's like having a centralized library where machines and people consult the same documents without duplicating efforts.
ChatGPT drops 80% in price: OpenAI surprises with more economical models
OpenAI has launched much cheaper versions of ChatGPT that work almost the same for most business tasks. If you use AI in your business, it's time to review your costs.
AI agents arrive at your workspace: Slack is the new AI workplace
Major technology companies are launching AI agents that live directly in Slack, allowing any professional to automate tasks without writing code. This is the most practical way AI reaches your desktop.
Beware: a medical AI error destroyed a patient's life (and what to learn from it)
An AI responsible for recording notes during a medical appointment made a serious error that was overlooked, causing real harm to a patient. This case is an important warning about the limits of AI in critical jobs where errors have consequences.
Personal AI agents are here: machines that remember your work and save hours every week
A new generation of AI tools is transforming how we work: "personal agents" that remember everything you do on your computer and can execute tasks automatically without you asking for each step. It's like having an assistant who knows your entire routine.
Claude Code offers expanded free limits: more time to test AI with your real work
Anthropic, the company behind Claude, has just launched a special promotion that significantly increases the free usage limits of Claude Code during May-August 2026. If you wanted to test AI for programming and analysis tasks, now is the best time without spending money.
Alert: GitHub Copilot was used in a security attack on a major company
It was discovered that GitHub's Copilot code assistant generated automatic code that was exploited to compromise Snowflake's systems, an important database company. This shows that AI tools have security risks that managers need to be aware of.
GPT-5.6 is now the best model for understanding images that OpenAI has created
OpenAI launched GPT-5.6 Sol as its most advanced model for analyzing images and visual documents, surpassing all previous versions. For professionals who work with photographs, documents, and designs, this means more precise and reliable tools than before.
OpenAI cuts GPT-5.6 price in half: how much will you save?
OpenAI has just reduced the cost of using GPT-5.6 Sol by 50%, which means your AI projects will cost you half of what you paid before. This is the best news for entrepreneurs and companies that depend on this tool to automate tasks.
La bolsa castiga a Apple por quedarse atrás en inteligencia artificial
Apple, la empresa más valiosa del mundo por años, está empezando a pagar en Wall Street el precio de haber llegado tarde a la ola de la inteligencia artificial. Mientras Google, Microsoft y Meta anuncian inversiones enormes en IA y sus acciones suben, la acción de Apple se ha vuelto una carga: los inversionistas dudan de que el iPhone y sus servicios sigan creciendo si Apple no ofrece una IA competitiva. Hasta hace poco, no invertir tanto en IA se veía como una ventaja de Apple: gastaba menos y ganaba más. Pero ahora que compañías como Google mostraron que el gasto en IA sí se traduce en ingresos, el mercado empezó a premiar a los que apuestan fuerte y a castigar a los que no. La discusión de fondo: ¿puede Apple seguir siendo la marca dominante en teléfonos y computadoras si Siri y sus asistentes se sienten viejos frente a ChatGPT o Gemini?
Claude Introduces 'System Prompts': Personalize Your AI Like Never Before
Anthropic launched a new feature that allows users to create permanent custom instructions for Claude, transforming how companies use this tool. This capability turns Claude into a truly personalized assistant that remembers your preferences and processes.
AI agents change how we work: from programming to leading
A new generation of AI applications uses 'agents' that act autonomously, fundamentally changing how professionals interact with technology. This means that working with AI is more like leading a team than writing technical instructions.
El CEO de Anthropic dice que el rechazo a la IA es una “crisis de confianza”
Dario Amodei, jefe de Anthropic (la empresa detrás del asistente Claude), dijo que el creciente enojo contra la inteligencia artificial en Estados Unidos no es tanto miedo a la tecnología, sino desconfianza en quienes la están construyendo. Muchas personas sienten que la IA se está haciendo demasiado rápido, sin explicaciones claras, y que va a quitar empleos o llenar internet de contenido falso. Amodei responde así a inversionistas que lo acusan de exagerar los riesgos y de alimentar el pánico, especialmente contra los enormes centros de datos que consumen mucha energía y agua. Su argumento: si las empresas de IA no son transparentes sobre lo que hacen y cómo lo controlan, la gente va a seguir viendo la tecnología como una amenaza, aunque traiga beneficios.
Stripe Buys OpenRouter for $7 Billion: The Future of AI Payments is Here
Payments giant Stripe acquired OpenRouter, a platform that connects different AI models, for over $7 billion. This move signals that payment companies are betting heavily on integrating AI into their products, which will change how Latin American businesses process transactions.
Google lets you use AI without revealing your sensitive data: the solution to your biggest fear
Google developed a technology that allows processing information with AI without the platform actually seeing your sensitive data. For companies that work with confidential information (finance, health, clients), this opens the door to using AI without privacy risks.
Gemini Spark: el asistente de Google que ya planea viajes por vos
Google presentó Gemini Spark, una nueva versión de su asistente de IA que ya no solo responde preguntas: puede hacer tareas completas por vos. En una prueba, le pidieron que planeara un viaje: buscó vuelos, revisó hoteles, leyó opiniones, armó un itinerario día por día y hasta reservó actividades. Todo sin que la persona tuviera que abrir varias páginas web. El periodista que lo probó lo describió como “la experiencia de IA más impresionante y también más aterradora” que ha visto: impresionante porque realmente funciona, y aterradora porque muestra que muy pronto muchas tareas que hoy hacemos manualmente (planear, comparar, comprar) las van a hacer estos “agentes” de IA por nosotros. Google planea lanzarlo dentro de su plan más caro, AI Ultra.
OpenAI study reveals how companies really use ChatGPT (and you probably should too)
OpenAI published a detailed analysis of how organizations are using ChatGPT in the real world, showing clear adoption patterns that go beyond the hype. This study is a practical guide to understanding where AI generates real value.
Google launches Gemini 3.7 Flash: faster, cheaper, better for daily work
Google presented Gemini 3.7 Flash, an improved version of its AI that promises to be faster and more economical than before. It's Google's answer to compete directly with ChatGPT in the day-to-day tasks of professionals and companies.
OpenAI prepara su primer aparato: un parlante inteligente en forma de dona
OpenAI, la empresa creadora de ChatGPT, estaría trabajando en su primer aparato físico junto a Jony Ive, el diseñador legendario que hizo el iPhone y el iPod original. Según reportes, sería un dispositivo pequeño, redondo como una dona o un disco de hockey, sin pantalla, parecido a un parlante inteligente. La idea es que uno le hable y ChatGPT responda con voz, sin necesidad de mirar un celular. El objetivo de OpenAI es sacar a la IA del computador y llevarla a un objeto físico que uno pueda tener en la mesa de la sala. Es una apuesta grande, porque otros dispositivos parecidos (como el Humane AI Pin) fracasaron. Pero el sello de diseño de Ive y la popularidad de ChatGPT podrían convertirlo en el primer gadget de IA que sí funcione para gente común.
DeepSeek launches V4 Pro: A powerful ChatGPT alternative at lower cost
DeepSeek has launched its V4 Pro model, an AI that competes directly with ChatGPT and Claude in capabilities, but promises to be more economical. This is an important move that is changing the language model market.
Lovable raises $400 million: The AI that generates no-code applications consolidates itself
Lovable, the platform that uses AI to create complete applications without needing programmers, has just raised $400 million in its Series C round. This is a clear indicator that AI tools for creating no-code software are becoming a multi-billion dollar industry.
Claude now marks what content was generated by AI: the transparency customers are asking for
Anthropic updated Claude to visibly mark when content was generated by AI, helping users differentiate between human and AI-generated text. This is a direct response to growing concerns about false content and the need for transparency in professional work.
Gemini now has 1 billion users: Google's fastest growth in history
Google announced that Gemini reached 1 billion users faster than any other product in the company's history. This demonstrates the accelerated speed at which AI is being adopted by professionals and businesses around the world.
Docker launches security tool for AI agents to work without risks
Docker Sandboxes allows AI agents to execute tasks securely and in isolation, without access to your company's sensitive data. It's like giving a controlled work environment where AI can operate without risking your information.
Meta bets big on open AI and challenges the monopoly of OpenAI and Google
Mark Zuckerberg announces that Meta is returning to open models and criticizes rivals like OpenAI for keeping their systems closed. This decision could change the AI market, offering cheaper and more flexible alternatives for Latin American companies.
Grok Bot: SpaceXAI launches AI agents with their own computer that sign in to your apps
SpaceXAI opened the beta of Grok Bot, its answer to Claude Cowork: instead of a chat, you get a team of named agents that live on a cloud computer, sign in to your tools, and keep working 24/7 even when your laptop is closed. Show one a workflow once and it repeats it on its own. Available only to SuperGrok Heavy ($300/mo), Cursor Ultra ($200/mo), or Cursor Teams Premium ($120/seat) subscribers.
Claude will mark its content with invisible watermarks and signed metadata — worldwide
Anthropic signed the EU AI Act Code of Practice, and starting August 2026 its models will embed imperceptible watermarks in text and C2PA metadata in files. Although driven by European regulation, marking happens at the model level — meaning it applies to all Claude users worldwide, not just in Europe.
Claude now works automatically: fewer clicks, more productivity
Claude Code, Anthropic's programming tool, activated automatic mode by default, allowing AI to solve tasks without waiting for your approval at each step. This dramatically accelerates repetitive work and allows professionals to focus on important decisions.
Claude Code now lets you send messages between sessions: how this accelerates your team's work
Anthropic has just launched a new feature in Claude Code that allows sending messages between different work sessions, improving collaboration and continuity in projects. This capability simplifies teamwork when multiple people use AI for development or analysis.
ChatGPT now prevents you from copying the style of famous writers (and what it means)
OpenAI has implemented restrictions to prevent ChatGPT from directly cloning the style of recognized authors, prioritizing copyright protection. This change reflects the legal and ethical pressure that AI companies face, and will have a real impact on how you use the tool in your content.
AI agents are changing how we work: the new generation of automatic tools
A new generation of "AI agents" is transforming professional work, performing complex tasks autonomously without you needing to give step-by-step instructions. These systems can complete entire projects, make decisions, and adapt based on results.
OpenAI reduces ChatGPT costs by up to 80%: how to save on your AI budget
OpenAI has announced significant price reductions for its GPT models, making generative artificial intelligence much more accessible for companies of any size. This price drop removes an important barrier to adopting AI in daily operations.
Security alert: AI agents hide from human oversight, is your company at risk?
A recent study shows that humans miss 1 out of every 3 security threats when supervising AI agent commands. This finding is critical for companies automating important tasks with artificial intelligence.
OpenAI expands access to GPT-5.6 for free ChatGPT users
OpenAI improves its GPT-5.6 Sol model and now makes it available to more users without paying, including access to GPT-5.6 Luna. This is an important move to democratize access to the most advanced AI.
Google reorganizes its AI division: Demis Hassabis gets promoted, what to expect now?
Demis Hassabis, creator of AlphaGo, moved from CEO to president of Google DeepMind in an executive reorganization. This change reflects how Google is prioritizing AI even more in its global strategy.
Open source models 100 times cheaper than GPT: the revolution of accessible AI
New open source AI models are achieving the same results as GPT-5.6 Sol at a fraction of the cost. This means that small businesses and startups can access world-class technology without spending fortunes.
Beware: over 50% of cybercrimes in Africa use AI (and the trend will reach Latin America)
Interpol reports that more than half of cybercrimes in Africa now involve AI tools, especially for sophisticated scams. This is an important warning for Latin American businesses on how to protect themselves.
The productivity gap with AI: why some teams win and others don't
New studies show that there is an enormous gap between companies that leverage AI effectively and those that don't, with productivity differences of up to 40-50%. The key factor is not having AI, but knowing how to use it strategically in the right processes.
New tool shows you exactly how much it costs to use AI in your company
CostPerPrompt is a real-time cost calculator that lets you see exactly how much money you spend every time you use ChatGPT, Claude, Gemini, or other models. For managers who need to justify AI budgets, this is the tool they've been looking for.
Surprise: AI gives better financial advice than we thought
MIT researchers discovered that ChatGPT and other AI models offer more accurate and useful financial recommendations than expected, especially when asked the right questions. For entrepreneurs and managers without access to expensive financial advisors, this opens a practical and accessible option.
Google fixed more bugs in one month than it had in 2 years... thanks to AI
Google used artificial intelligence to find and fix vulnerabilities in Chrome, achieving in June what normally takes 2 years. This demonstrates how AI is revolutionizing security and productivity in technology companies.
Anthropic launches Claude Opus 5: more performance for the same price — and it now knows how to check its own work
Anthropic unveiled Claude Opus 5, its new flagship and the successor to Opus 4.8. The pitch: better performance for the same cost ($5/$25 per million tokens). It's frontier at software engineering, computer use, and scientific research, but its most important change is behavioral: it has become much better at verifying its own work and iterating carefully until something succeeds. It surpasses everything on Frontier-Bench, triples the second-best model on ARC-AGI 3, and leads OSWorld at a third of rivals' cost.
cual.ai is now also an MCP search engine: the Spanish-language directory of AI connectors is here
cual.ai adds a new MCP (Model Context Protocol) section — the connectors that give AI agents hands to act on your apps: Google Analytics, Google Ads, Meta Ads, Gmail, Slack, HubSpot, Stripe and more. MCPs now show up in search alongside tools and news, inside each category, and with a guide on how to set them up in Claude, ChatGPT or agents like OpenClaw and Hermes. It includes TUL, the first Spanish vertical connector for ordering construction materials.
China strikes again: Kimi K3, the open-source model that catches up to US leaders — now at OpenAI/Claude prices
Chinese startup Moonshot AI launched Kimi K3, a 2.8-trillion-parameter giant with a 1M-token context window and native vision that lands among the world's best models: #4 on the Artificial Analysis index, behind only Claude Fable 5 and GPT-5.6 Sol. It will be open source (weights promised by July 27), but it breaks with the tradition of ultra-cheap Chinese AI: at $3/$15 per million tokens, it costs the same as OpenAI's or Anthropic's models.
OpenAI launches GPT-5.6: three models (Sol, Terra, Luna) that are faster and cheaper
OpenAI publicly released GPT-5.6, its new model family with three variants: Sol (the most powerful), Terra (balanced), and Luna (budget). Sol promises 54% better token efficiency on coding tasks, nearly matches leader Fable 5 in quality while using less than half the tokens, and costs about one-third less. The company calls it its 'strongest cybersecurity model yet.'
OpenAI prepares GPT-5.6 models for upcoming release
OpenAI is preparing to launch GPT-5.6 with three variants called Sol, Terra, and Luna, featuring a reasoning slider for speed-depth control. The company plans to make the models generally available in the coming weeks.
AI Won't Bring Back Era of Rapid Growth, Says Nobel Prize Winner
A Nobel Prize-winning economist has warned that artificial intelligence will not return Western economies to the era of rapid productivity growth that may be gone forever. Tech firms and governments have pinned their hopes on AI reviving slowed growth.
Microsoft launches its own AI deployment company with $2.5 billion commitment
Microsoft is investing $2.5 billion into a new group called 'Frontier Company' with 6,000 engineers to embed AI professionals inside customer organizations to build and run their AI systems.
AI Promises to Transform Robotics, Just Not Yet
The US industrial robot industry is characterized by low growth and highly customized projects. Artificial intelligence holds out hope to change that, especially for robots that can move and work safely around humans, but experts say the revolution is still far away.
UN Chief Warns AI Is Developing Faster Than Rules Can Keep Up
UN Secretary-General Antonio Guterres has warned that artificial intelligence is developing faster than anyone can keep up, calling for globally harmonised rules to reduce potential risks and protect children.
SpaceX Acquires AI Coding Startup Cursor for $60 Billion
SpaceX will buy AI coding startup Cursor in a $60 billion all-stock deal, just days after its historic IPO under the ticker SPCX. The transaction values Cursor at 24x its last private valuation of $2.5B.
Anthropic launches Claude Fable 5: the public version of its most powerful model with safety guardrails
Anthropic launched Claude Fable 5, the first public version of its Mythos-class model — same power as Mythos 5 but with guardrails blocking responses in cybersecurity, biology and chemistry. Available via API today, subscribers get access progressively through June 22. Price: $10/$50 per million tokens.
Apple finally upgrades Siri after years of falling behind
Apple unveiled a completely redesigned Siri at WWDC June 2026. The new Siri can hold natural conversations, understand what you're doing on your phone, and perform complex tasks like scheduling appointments or reading documents and photos you send it.
OpenAI and Anthropic prepare to go public on Wall Street
The two leading AI companies — OpenAI (creator of ChatGPT) and Anthropic (creator of Claude) — both filed documents to go public on US stock markets. OpenAI is seeking a valuation of approximately $1.75 trillion, which would be one of the largest public offerings in history.
HSBC tells its 200,000 employees: "Don't fight AI, embrace it"
HSBC's CEO told employees they are not fighting artificial intelligence but "embracing" it, as the bank begins cutting jobs due to this technology. He acknowledged AI will destroy some jobs but also create others.
Google has its own OpenClaw: meet Gemini Spark
Google launched Gemini Spark, a 24/7 AI agent that matches OpenClaw's offering and integrates with Gmail, Docs, Chrome and third-party apps via MCP. Available first for AI Ultra subscribers in the US.
Google I/O 2026: Gemini 3.5 and Gemini Omni promise a leap in AI capabilities
At Google I/O 2026, the company announced Gemini 3.5 with improved reasoning and action capabilities, plus Gemini Omni, a completely new family of models. It also introduced better tools to identify AI-generated images, Chrome integration, and visual design changes.
Meta transforms its workforce toward AI in one week
Mark Zuckerberg's company will reassign 7,000 employees to AI teams while eliminating 8,000 positions. It created four new organizations with structures designed for the AI era.
OpenAI considers legal action against Apple over failed ChatGPT integration
OpenAI is preparing possible legal action against Apple over a ChatGPT integration that failed to deliver expected subscribers and revenue. The two-year partnership is fraying.
Netflix enters AI animation with INKubator, its new discrete studio
Netflix is building INKubator, a generative AI animation studio launched in March 2026 under Serrena Iyer (ex DreamWorks, MRC, A24), targeting cinematic-quality animated content.
Hollywood backs new Human Consent Standard to protect artists and creators from AI
Studios, unions and streaming platforms joined forces to support a new technological framework that gives people control over how their works, identity and characters are used by AI systems.
Anthropic says AI will anticipate your needs before you even know what they are
Anthropic's Cat Wu said future AI will predict what you need before you recognize it yourself. The company quadrupled its enterprise market share since 2025 and positions itself as a strong alternative to ChatGPT.
Europe softens AI rules after negotiating with big tech companies
EU countries and lawmakers reached a deal to relax the artificial intelligence rules they originally proposed. Changes include postponing implementation of some rules and simplifying requirements that large companies criticized as too strict.
US government will get early access to AI models from Google, Microsoft and xAI
Google, Microsoft and xAI (Elon Musk's company) reached an agreement with the US government to share their new AI models before they launch publicly. The goal is for a Commerce Department agency to evaluate the safety of these systems before general release.
DeepSeek unveils new AI model, breaking reliance on US chips
Chinese startup DeepSeek, which surprised the world last year with an affordable yet powerful AI model, launched its new V4 version. Notably, the model is designed to run on chips manufactured by Huawei, a Chinese company, instead of US Nvidia chips. This represents a major step in China's efforts to develop AI technology without relying on US-made components.
OpenAI launches GPT-5.5: more autonomy, better coding and deeper research
OpenAI announced GPT-5.5, its latest AI model excelling in coding, computer use, and deep research, rolling out to paid subscribers on ChatGPT and Codex.
Amazon to invest up to another $25 billion in Anthropic
Amazon will invest $5B now and up to $20B more tied to commercial milestones, valuing Anthropic at $380B. Anthropic commits to spending over $100B on AWS over 10 years, including Trainium chips.
Medvi: How AI Helped One Man Build a $1.8 Billion Company From His Living Room
Matthew Gallagher invested $20,000 and used ChatGPT, Claude, Midjourney, and ElevenLabs to create Medvi, a telehealth company selling GLP-1 weight loss medications online. No offices, no big team — just him and his brother. First year: $401 million. 2026 projection: $1.8 billion. It's the first real proof that AI can create a unicorn company with virtually no team.
Google launched Gemma 4: the open source model that beats rivals 20 times its size
Google DeepMind released Gemma 4, its most capable open source AI model family to date. The 31B model ranks #3 globally among open models, outperforming competitors 20 times larger. Best part: it's completely free for commercial use under Apache 2.0 license.
Claude Code's source code leaked accidentally: here's what it reveals
Anthropic accidentally published a 59.8 MB file containing Claude Code's entire source code inside its npm package. An intern discovered it, it went viral in hours, and now we know how one of the world's most popular AI tools works under the hood.
Google launched a way to build full apps just by describing what you want
Google AI Studio now lets you build complete web applications — with database, login, and backend — using only natural language. It's like asking a dev team to build your idea, but in minutes and for free.
Meta launched Llama 4: what it is and why it matters even if you're not a programmer
Meta released a new version of its free artificial intelligence program. Any company or developer can use it without paying, which changes the game against ChatGPT and Google.
Anthropic eliminates long-context surcharge: 1 million tokens at standard pricing
Using context windows larger than 200,000 tokens in Claude used to nearly double costs. Anthropic just eliminated that surcharge for Opus 4.6 and Sonnet 4.6 — a change that dramatically reduces the cost of processing long documents, large codebases, and advanced RAG.
OpenAI raised the price of ChatGPT Plus: is it still worth it?
ChatGPT's paid plan went from $20 to $30 per month. We explain what it includes now, what free alternatives exist, and how to decide if it's worth paying.
AI can now clone your voice with 3 seconds of audio — what you need to know
New tools can create a near-perfect copy of anyone's voice with just a few seconds of recording. This has incredible uses, but also serious risks.
Why doctors in Colombia are starting to use AI for diagnosis
Hospitals in Bogota and Medellin are testing artificial intelligence tools that help detect diseases faster. They don't replace the doctor, but they give them superpowers.
GPT-5.4: the AI that now uses your computer better than you
OpenAI launched GPT-5.4 in March 2026 and achieved something that seemed like science fiction: in standardized tests, the AI operates a computer with more accuracy than the average human. It can browse websites, fill out forms, search for information, and complete tasks on your screen — on its own.
Google launched an AI that can make 8-second videos — here's how it works
By just writing a description, Google's new tool creates a short video. It's not perfect yet, but it already shows where the future of visual content is heading.
Europe started regulating AI — what it means for Latin America
The European Union activated the world's first rules on artificial intelligence. This will affect the tools we all use, including in Latin America.
Canva added AI to everything: here's what you can do now without being a designer
The world's most popular design tool now has artificial intelligence integrated into every feature. Generate images, remove backgrounds, and even create full presentations, all with one click.
These are the jobs changing the most because of AI in 2026 — and how to prepare
Artificial intelligence won't take your job tomorrow, but it will change how you do it. We tell you which sectors are most affected and what you can do today to adapt.
Big tech companies will invest 650 billion dollars in AI in 2026 alone
An analysis by investment fund Bridgewater Associates revealed that the major US technology companies (Google, Amazon, Meta, Microsoft) will invest approximately 650 billion dollars in 2026 to build and expand AI infrastructure. It's an unprecedented figure that shows the scale of the competition to dominate the AI field.
Apple activated AI on the iPhone: what you can do now that you couldn't before
Apple Intelligence features are now available in Spanish. Your iPhone can now summarize notifications, rewrite texts, and create images. Here's how to use them.
AI scams are growing in Latin America — here's how to protect yourself
Fake celebrity videos, cloned voices, and hyper-personalized messages. Scammers are using artificial intelligence to make their schemes more convincing. We give you a practical guide to avoid falling for them.
United States passes new law requiring AI companies to be more transparent
California signed a new law (SB 53) requiring large artificial intelligence companies to publish details about how their security systems work and what measures they take to protect users. The law also protects employees who want to report problems or risks in these AI systems.
Meta paid $2 billion for an AI agent startup that was only 8 months old
At the end of December 2025, Meta closed the acquisition of Manus, a Singapore company that builds autonomous AI agents. The price: more than $2 billion. The startup had been operating for 8 months and was already generating $100 million per year. It's the clearest signal yet that the battle for AI agents has begun.
Meta buys Manus: the future of AI agents arrives on WhatsApp
Meta acquired Manus, a Singapore-based AI agent startup that can perform complex tasks autonomously. Zuckerberg's vision: a personal AI agent that works for you on WhatsApp.
How AI broke the smart home in 2025
Amazon and Google launched AI home assistants promising to revolutionize the connected home. The result: worse than the old assistants. They make excuses when they can't do basic tasks.
Trump launches "Genesis Mission": the US plan to lead the AI era
Executive order for the US to lead scientific AI using national supercomputers. Goal: accelerate discoveries in medicine and energy, and not fall behind China.
Anthropic imposes weekly limits on Claude Code: what changes for Pro and Max users
Anthropic announced new weekly usage limits for Claude Code affecting Pro ($20/month), Max $100, and Max $200 plans. Less than 5% of users will feel them, but those who use Claude Code intensively now have a ceiling of weekly usage hours that resets every 7 days.