Meta Advances Orion AR Platform for Next-Gen Computing

Meta unveiled Meta Orion AR glasses — a prototype AR glasses platform developed over 10 years by Reality Labs to connect the physical and the virtual.  

Here are the main features of Meta’s Orion AR Platform: 

Design And Display Technology 

  • Meta Orion AR glasses feature clear lenses, enabling users to see their surroundings, make eye contact, and interact with others, unlike large mixed reality headsets that rely on video pass-through.  
  • The Meta Orion AR glasses weigh under 100 grams and are made of magnesium, providing comfort for everyday wear.  
  • Meta Orion AR glasses project 3D holograms into the lenses using micro-LED projectors in the arms, giving a 70° field of view. The silicon carbide lenses help project light and produce bright, high-resolution images in multiple lighting conditions.  

Computing And Control System 

  • To keep the glasses light, a small, wireless device you carry in your pocket handles most of the processing, not the glasses themselves.  
  • You mainly control Meta Orion AR glasses with an EMG wristband that reads electrical signals from your wrist to detect small hand movements. This layered approach lets you wipe, click, or scroll without moving your arm visibly.  
  • Multimodal interaction: Meta Orion AR glasses combine voice, eye, and hand tracking to deliver a fluid, intuitive user experience.  

Key Capabilities And AI Integration 

  • Meta AI on Orion recognizes what you are looking at and offers helpful suggestions, such as ingredient identification and recipe recommendations.  
  • The glasses display several apps at once, like a movie screen for videos, a web browser, or social media feeds.  
  • You can take part in video calls where others appear as holograms in the room using Meta Orion AR glasses.  

Availability And Future Outlook 

  • Currently, Meta is testing internally with employees and a few external groups. The product is not yet available for sale to the public.  
  • Meta plans to eventually sell Meta Orion AR glasses at a price similar to a high-end smartphone or laptop.  
  • Impact on strategy: The custom silicon and EMG technology developed for Orion is already shaping other Meta products, such as the Ray-Ban Meta glasses and the Quest headset. This is helping Meta move forward into an era of always-on computing.  

Today at Connect, Mark Zuckerberg introduced Meta Orion AR glasses, our first true AR glasses. They were previously known as Project Nazare.  

Orion offers users an industry-leading field of view for wider, more immersive visuals. Its silicon carbide lenses, advanced waveguides, and ULED projectors deliver sharper, brighter images. Orion provides daily utility and an enhanced user experience, making it our most advanced and polished prototype to date.  

We are focused on making Orion more accessible by reducing costs and improving performance, so more people can benefit from powerful AI. They are experienced. Our goal is to help people interact with information and their surroundings in a seamless, transformative way.  

“We are building AR glasses.”  

Five years ago, we said those five simple words. That moment signified our devotion to a future in which people do not have to choose between digital information and the physical world.  

Now, five years later, we have another five words that could change everything again:  

At Meta, our mission is to give people the power to build community and bring the world closer together. Reality Labs creates tools that help people feel linked wherever they are. We are working on the next computing platform to put people first so they can feel more present, connected, and empowered.  

Ray-Ban Meta glasses demonstrate the convenience of hands-free access to digital life. Users can talk to an AI assistant, connect with friends, and capture special moments easily, making daily tasks more efficient and enjoyable — all without reaching for their phone.  

Ray-Ban Meta created a new category of AI-powered glasses without a display, meeting demand for simple wearable tech. The XR industry’s ultimate goal is true AR glasses with a large augmented display and personal AI help for richer, more immersive everyday experiences in a stylish, all-day-wear design.  

Today, Orion brings the dream of advanced AR closer by offering immersive digital overlays and seamless interactions, the result of years of innovation. Reality Labs has produced custom silicon and the most advanced Air display, enabling users to enjoy more powerful experiences on lightweight glasses rather than bulky headsets.  

Orion’s intuitive input system voice, eye movement, hand tracking, and an EMG wristband allows users to swipe, click, and scroll with minimal effort. Stay engaged with people around you while blending digital content into your physical world, building on these advances. We are taking the next step.  

Starting today at Connect and continuing through the year, we are giving Meta employees and selected external groups access to the Meta Orion AR glasses prototype. This step will gather feedback, enable further improvements, and support our commitment to launching a consumer AR glasses product line in the near future.  

Why AR Glasses? 

AR glasses matter for three reasons, driving the next step in human-centered technology.  

  • Meta Orion AR glasses enable digital experiences beyond the limits of a small smartphone screen. Augmented reality displays use your surroundings to place 2D or 3D content, anywhere you like.  
  • They come with built-in AI, which means artificial intelligence — a system that senses and understands your environment. This lets the glasses anticipate and assist with your needs. They work well both indoors and outdoors.  
  • Our industry’s goal has been to combine variable convenience, a large display, fast input, and smart AI in a comfortable everyday design. All in a design that people feel comfortable wearing. We’ve had to choose between bulky headsets for immersive experiences and lightweight glasses that can’t deliver rich apps because they lack a large display and computing power.  

We want all the benefits without compromise. We’ve worked to shrink VR and MR technology into lightweight, stylish glasses, designing, building displays, creating AR experiences, and inventing interactions all in one product and this has been our industry’s toughest challenge. Sometimes we thought our chances of success were below 10%.  

Until now.  

A Cutting Edge Display in an Exceptional Form Factor 

Orion offers about a 70-degree field of view (how much of your surroundings you can see through your lenses), the largest ever in such a small pair of AR glasses. This wide view makes it possible to use Orion for things like multitasking, watching big‑screen entertainment, or seeing life‑size projected people. All this digital content blends smoothly with what you see in the real world. A wide field of view was our top goal. We had to work against the laws of physics to bend light in new ways, all while keeping power use extremely low, measured in milliwatts.  

Instead of glass, we used silicon carbide for the lenses, a new material for AR glasses. Silicon carbide is very light, avoids optical artifacts (visual errors like glare or ghost images) or stray light, and has a high refractive index (a measure of how much the material bends light) – These features are important for creating a wide field of view. The wave guides (thin structures that direct light) have complex nanoscale 3D structures to spread light as needed. These projectors use U LEDs (microscopic light-emitting diodes), a new display technology that is very small and uses little power. It looks like a regular pair of glasses with clear lenses. Unlike MR or other AR glasses, you can still see each other’s real eyes and expressions, so you can be present and share experiences. We needed dozens of innovations to make the design as comfortable and modern as everyday glasses. Orion is a real achievement in miniature, with components packed into tiny spaces. We even fit seven small cameras and sensors into the frame rims.  

We had to maintain optical precision at just one-tenth the thickness of a human hair. The system can detect even smaller movements, such as the frames expanding or shrinking with changes in room temperature, and quickly corrects the optical alignment in milliseconds. We used magnesium for the frames, the same material found in F1 race cars and spacecraft, because it is light but strong. This helps keep the optical parts properly aligned and efficiently carries heat away (which helps prevent the glasses from becoming too hot to wear).  

Heating and Cooling  

After we solved the display and zigzag challenges, we faced another big hurdle: combining powerful computing with low power use and effective heat management. Since you can’t put a fan in a pair of glasses, as you can with today’s MR headsets, we had to find new solutions. Many of the materials we use to cool Orion are actually similar to those that NASA uses to cool satellites in space.  

We designed custom silicon that is very power-efficient and designed for our AI, machine perception, and graphics algorithms. We created several custom chips, each with many specialized silicon IP blocks. This allows us to run complex algorithms for hand and eye tracking, as well as SLAM, using only a few dozen milliwatts of power, rather than the hundreds typically required. This means much less heat is produced.  

Together, these breakthroughs mark a turning point for Orion. With custom silicon at its core and continuous innovation, we’re not just shaping AR technologies—we’re charting the future of personal computing.  

Effortless EMG 

Each new computing platform changes how we interact with our devices; the mouse made graphical user interfaces possible, and smartphones only became popular once touch screens arrived. The same pattern is happening with wearables.  

We have discussed electromyography, or EMG, for years because we believe AR glasses require input that is fast, user-friendly, reliable, subtle, and seamlessly integrates into daily life. This technology is now ready for everyday use.  

Orion’s input system combines voice, eye, and hand tracking with an EMG wristband, so you can swipe, click, and scroll easily.  

It works and fails almost like magic. Picture taking a photo during your morning jog with just a tap of your fingers or moving through menus with tiny hand movements. Our wristband uses high-performance fabric with built-in EMG sensors to pick up even the smallest muscle signals. An on-device ML processor reads these signals and turns them into input events, which are wirelessly sent to the glasses. The system learns from you, so it gets better at perceiving subtle gestures over time. Today, we are also sharing more about how we support outside research to make EMG wristbands more accessible and fair.  

Meet the Wireless Compute Puck 

Real AR glasses must be wireless and compact. Our wireless compute puck handles some processing, helping Orion last longer and stay small and wireless, complete with low latency.  

  • The glasses handle hand tracking, eye tracking, SLAM, and special AR graphics, while the app logic runs on the puck. This setup keeps the glasses light and compact.  
  • The puck uses two processors, including one custom-made at Meta. This gives it the power needed for fast graphics, AI, and extra machine perception features.  
  • Because it’s small and sleek, you can easily put the puck in your bag or pocket and go about your day with no hassle.  

AR Experiences 

Like any hardware, what matters most is what you can do with it; even though it’s early or young, it already gives a promising look at the future.  

Meta AI, our smart assistant, runs on Orion. It can see what you’re looking at and help with helpful visualizations. Orion uses the same Lama model as the Ray-Band Meta, smart glasses. It also uses custom research models to show what future variables could do.  

You can make hands-free video calls on the go to catch up with friends and family in real time. Stay connected on WhatsApp and Messenger to view and send messages. There’s no need to take out your phone or search for apps. You can do it all through your glasses.  

You can play shared AR games with family far away or with a friend sitting next to you. Orion’s large display also lets you multitask with several windows, so you can get things done without carrying your laptop. Help chart the roadmap for our consumer AR glasses line. Our teams will continue to iterate and build new immersive social experiences alongside our developer partners, and we can’t wait to share what comes next.  

A Purposeful Product Prototype 

Orion won’t be sold to customers, but it’s not just a research prototype. It’s the most polished product prototype we’ve made and could be reproduced for the market. Instead of rushing it out, we’re focusing on internal development to keep improving the technology and experiences.  

This approach will help us deliver an even better product to consumers faster.  

What Comes Next 

There have been two big challenges for mainstream AR glasses: making a large display fit into compact glasses and creating useful AR experiences that work on them. Orion is a major step forward, offering real AR experiences on stylish hardware for the first time.  

Now that Orion has been introduced, we’re focusing on a few key areas:  

  • Tuning the AR display quality to make the visuals even sharper.  
  • Making the design even smaller wherever possible  
  • Producing at scale to lower the cost  

Over the next few years, you will see new devices from us that build on our research and development. Many of Orion’s innovations are already in our current products and future plans. They’ve improved our spatial perception algorithms. They now run on both Meta Quest 3S and Orion. The eye-gaze and gesture input system originally developed for Orion will be used in future products. We are also looking into using EMG wristbands in upcoming devices.  

Orion is more than a glimpse of the future; it shows what’s possible right now. We built it to help people connect, which is what we do best. From Ray Band, Meta Glasses to Orion, we’ve seen how these tools help people stay present and empowered in the real world. They also help people enjoy the benefits of the digital world.  

We believe you shouldn’t have to choose between the physical and digital worlds. What’s with the next computing platform you won’t have to orient is paving the way for a future where technology seamlessly empowers people to connect, create, and experience more without compromise. 

Source: Orion: True AR Glasses Have Arrived 

Anthropic Advances Enterprise AI Agents For Workplace Automation

Corporate productivity is changing as companies move from traditional software to increasingly dynamic agent-based systems. Anthropic has expanded its expert services to include autonomous agents built for complex workplac/+ automation. Instead of just handling simple conversations, these digital assistants- can search internal databases, work with third-party apps, and complete multi-step tasks with little human help. By focusing on a safety-oriented design, Anthropic addresses key enterprise concerns, including data control, reliability, and transparent record keeping. As businesses face digital transformation, these advanced agents help reduce administrative work and speed up decision-making.  

The Engineering of Functional Agency 

The key to this progress is a powerful reasoning engine that lets agents do more than just observe patterns they can focus on completing real tasks in the past. Workplace automation relied on strict rules that often failed with complex business data. Anthropics’ new agents use a set of guiding principles called constitutional design to understand uncertain instructions and make logical decisions. For example, if someone requests a quarterly audit, the agent finds the relevant data, identifies any financial issues, and produces a clear report. This independence stems from a system that allows the agent to start specialized processes to handle different parts of a complex task.  

This move toward functional agency is made possible by new tool-use protocols—standards that allow agents to interact directly with business software. These agents work directly with a company’s software, such as CRM systems, ERP software, and cloud storage, using standard APIs (application programming interfaces). Agents can update client records in Salesforce or create invoices in SAP. This makes the agent a link between different software systems, helping information flow smoothly between divisions without requiring people to transfer data by hand.  

Strengthening Operational Reliability and Safety 

One major challenge with autonomous systems is the black box problem, where understanding how a machine made a decision can be difficult. To address this, Anthropic added a chain-of-thought transparency layer to its enterprise agents. A transparency layer is a protocol in which, for every action, the agent builds a private logic trace a step-by-step record showing its reasoning, the data it used, and the rules it followed. This allows IT and compliance teams to review the agent’s work in real time, ensuring it follows company policies and legal standards. In areas like legal or health care, this traceability supports trust.  

The safety system also includes a strong human-in-the-loop setup. Companies can set certain important events, such as transferring money or deleting sensitive data, that require a human supervisor’s final approval. When the agent reaches one of these points, it stops and presents the supervisor with a summary of what it plans to do and the evidence supporting it. In this teamwork, while the agent handles most of the data work, humans still make the key decisions. This approach reduces the risks of full automation while retaining the speed and capability benefits.  

Improving the Modern Supply Chain 

These agents are especially valuable in managing complex supply chains in today’s changing environment, real-time response is a key advantage. Anthropic’s agents monitor logistics needs and feed weather and supplier data simultaneously. If a port in East Asia closes, the agent identifies affected shipments, gauges production impacts, and suggests alternate routes for our suppliers. This proactive approach lets supply chain managers focus on strategy over crisis response, preserving continuity during disruptions.  

The agents also optimize inventory by analyzing past sales data and current market trends, and forecasting demand spikes. The agent adjusts procurement orders to avoid overstock or shortages. This precision reduces waste and lowers the carbon footprint, boosting business efficiency and protecting the environment. In retail, this ensures the right product is in the right place at the right time, increasing customer satisfaction and reducing excess inventory costs.  

The Transformation Of The Knowledge Worker 

As these agents assume repetitive, data-intensive tasks, the role of the knowledge worker is elevated. Automating administrative functions such as scheduling, data entry, and basic document drafting enables experts to focus on strategy, problem-solving, and relationship-building. For example, in a law firm, a junior associate can use an agent for initial document review, freeing up time to develop legal arguments and advise clients. This shift improves the value of human workers, positioning them as pioneers in automation rather than routine practitioners.  

This transformation also benefits professional development. Anthropic’s agents serve as knowledge co-pilots, giving employees immediate access to organizational knowledge. If a new engineer encounters problems with a traditional codebase, the agent can offer a narrated walkthrough that references design documents and past bug reports. This reduces onboarding challenges and ensures that institutional knowledge is preserved and accessible company-wide. By making information readily available, the agents promote constant education and agility, both of which are essential in today’s fast-changing economy.  

The Quiet Architecture of the Future 

As digital technology becomes integral to our work lives, we are quietly refining our methods. The Office of the Future will embrace smart, unseen systems that simplify tasks and drive efficiency. These tools will align with our ambitions, safeguarding data and boosting productivity. As the boundary between human-machine tasks and human creativity fades, we gain more freedom to focus on ideas. Ultimately, technology will handle routine work, empowering us to explore new possibilities within a dependable, logical work environment.

Source: What 81,000 people want from AI 

Big Tech Invests in New Energy Solutions to Power AI Growth

Major technology companies such as Microsoft, Amazon, Google, and Meta are investing in new power systems. They aim to address the significant power needs of AI data centers. As AI workloads increase and place greater demands on electric grids, these firms pursue an integrated strategy. This strategy combines renewable energy with investments in nuclear power and natural gas to ensure continuous reliability.  

Key Investments and Strategies: 

  • Nuclear Energy (SMRs and Existing Plants): technology companies have invested in small modular nuclear reactors (SMRs) and existing nuclear infrastructure. Infrastructure to deliver carbon-free, high-capacity baseload power.  
  • Amazon: The company is investing over $500 million in nuclear development, has entered a $650 million agreement for power from a Pennsylvania plant, and is working with Energy Northwest to fund four SMRs.  
  • Microsoft signed a 20-year agreement in 2024 to restart the decommissioned Three Mile Island nuclear plant and partner with fusion startup Helion.  
  • Google signed an agreement with Kairos Power to build seven SMRs, with the first expected to be online by 2030.  
  • Meta: The company is collaborating with utilities to secure new nuclear energy for its data centers, including a reported 20 T contract with Conservation Energy.  

In addition, two nuclear energy companies are also turning to natural gas and hybrid energy systems to maintain 24/7 reliability. Tlass’s demand grows.  

  • Meta: The company is developing a large data center in Louisiana supported by three new gas plants. It announced plans for a gigawatt-scale data center powered by three small modular nuclear reactors.  
  • Combined strategy: Companies are integrating on-site renewable energy sources, such as solar photovoltaic and wind turbines, with battery energy storage systems and natural gas turbines to ensure a continuous electricity supply.  

Renewable Energy Expansion 

  • Microsoft: contracted over 34 GW of carbon-free electricity in 24 countries, including a $6.2 billion agreement in Norway to set the government’s target to 100% renewable energy.  
  • Amazon: The company met its goal of matching 100% of electricity usage with renewable energy seven years ahead of schedule in 2023 and continues to support new projects. Evolving energy strategies reflect a pivot to new sources. Here is why this shift is underway. The rapid growth of AI is slowly training the electric grid, with data centers expected to consume as much power as 100,000 homes. These initiatives help prevent grid bottlenecks, meet net-zero commitments, and ensure the reliable operation of energy-intensive air training.  

The rise of AI poses a unique opportunity to fuel economic growth, increase productivity, and support the community changes needed for the energy transition. At the same time, energy remains a top priority for policymakers and business leaders because it connects economic opportunity, innovation, industrial growth, digital transformation, and environmental impact.  

The Asia-Pacific region, with its rapidly growing economies, large urban populations, and dynamic manufacturing labs, is expected to account for two-thirds of global electricity demand by 2030, according to the International Energy Agency (IEA). This increase parallels a spike in digital infrastructure, including data centers, cloud computing, and artificial intelligence (AI).  

In 2024, Asia installed over 413 gigawatts (GW) of new renewable power capacity, representing 71% of the world’s total additions. In Southeast Asia, electricity demand is projected to triple by 2050, driven by demographic growth, urbanization, and rising air-conditioning demand. Expanding carbon-free electricity generation, upgrading transmission and distribution networks, and making AI infrastructure more sustainable will support energy security, job creation, and advancement in clean energy and economic development.  

Energy and Sustainability Solutions 

Across Asia, Microsoft is signing long-term agreements to source carbon-free electricity and deploying technologies such as AI-driven grid forecasting models and circular data center processes to enhance energy system reliability, efficiency, and sustainability. The company is also partnering with governments, utilities, and industry associations to advance policy reform and expedite progress. Initiatives include clean energy development and market facilitation, legislative advocacy, technology innovation, water resilience, and circularity all of which contribute to the energy transition.  

  • Expand Clean Energy Supply and Market Development: demand for carbon-free electricity is essential to expanding the clean energy supply. Microsoft has contracted over 34 GW of carbon-free electricity across 24 countries, including 19 GW in 2024, and is applying this approach in the Asia Pacific to enable developer financing and create expandable solutions for buyers. Recent agreements include a 20-year virtual PPA with Shizan Energy in Japan for 25 MW of rooftop solar, a portfolio with EDP Renewables in Singapore for up to 200 MW, and a 10-year agreement with Contact Energy for Hookah Unit 3 Geothermal Power Station in Aotearoa, New Zealand. Through our climate innovation fund, we support novel financing models that accelerate the adoption of clean energy in Asia. We have invested in Eversource Capital, which has mobilized $2,000,000,000 and avoided 13.4 million tons of CO2, and in SEACEF, which reduces risk for early-stage renewable projects in Southeast Asia. These investments help unlock private capital and speed up project delivery. Supporting policies and technologies that expand carbon-free electricity and grid infrastructure is also critical. Advocacy matters because government policies and regulations determine the pace of renewable growth, the affordability of clean power, and access for corporate buyers in Korea. Microsoft collaborated with PS suppliers and the Association on the Special Act on Expanding the National Power Grid, passed in February 2025. The act will strengthen Korea’s transmission system, enable greater integration of renewable energy, and create opportunities for corporate buyers. This demonstrates how collaborative advocacy and government leadership can remove barriers. Microsoft is also partnering with energy companies in Asia to use AI for clean power. One developer is building an AI platform on Azure to improve predictions of solar and wind output. Early results show better forecast accuracy, fewer costly errors, and more efficient maintenance of renewable energy even before new power lines are constructed.  
  • Strengthen Water Resilience: Water stress is a critical sustainability challenge in Asia, ranging from shortages and poor water quality to climate-driven variability. Data centers depend on water for cooling. So we are committed to reducing our water use and supporting community adaptation. Microsoft is investing in solutions that strengthen local water resilience and support our goal to be water-positive in Malaysia by 2030. We partnered with CLEAN International to install rainwater-harvesting and filtration systems in 50 schools, benefiting 20,000 people. In India, collaborations with Flux Gen and Botanical Water Technologies conserve millions of liters annually and provide portable water to underserved communities in Korea. Our partnership with K-Water will create a wetland to restore water flows equal to the daily needs of one million people.  
  • Advanced Circularity: Microsoft is also reducing data center waste. At the circular center in Singapore, decommissioned servers and cloud computing hardware are reused or recycled. Parts are distributed to schools, training programs, and manufacturers. Microsoft is investing in companies like Cyclic Materials, which recycle rare earth magnets from old equipment. This reduces the need for new mining and improves supply chain sustainability. 

AI As a Tool for Decarbonization 

AAI’s influence goes beyond the digital group when used effectively. AI can drive decarbonization across entire economies. The IEA estimates that widespread use of existing AI technologies could reduce global emissions. This could be up to three times the indirect emissions generated by data centers worldwide. 

Building on this global perspective, it is important to consider Asia’s unique context, where energy-related systems are complex and fast-changing. AI offers three core strengths:  

  • Measuring, forecasting, and optimizing complex systems such as national electricity grids in real time while balancing fluctuating renewable generation with electricity demand.  
  • Accelerating progress in materials science, such as new battery chemistries and low-carbon fuels.  
  • Empowering the workforce and decision makers by transforming fragmented datasets into practical insights that accelerate deployment, strengthen supply chains, and unlock financing.  

A strong example is AI’s application in grid management across Asia. Advanced forecasting tools can model renewable output, predict demand increases, and optimize the use of existing transmission assets. This allows more clean energy to connect without waste while new lines are being built.  

We are also exploring how AI can accelerate permitting reform, which is still a significant barrier to energy project deployment and development. Microsoft is partnering with organizations such as Idaho National Laboratory in the US and Lloyd’s Register in the UK to simplify energy project development and permitting, thereby reducing costs and time in Asia. We are in early discussions on how AI can support faster approval of renewable projects and grid interconnections.  

Policy And What Comes Next 

Policy acts as a multiplier when supported by supportive frameworks. Governments can unlock investment, accelerate innovation, and scale solutions more rapidly than any single company. Three imperatives stand out:  

  1. Expand grid infrastructure, address grid bottlenecks, and mitigate risks to digital growth and decarbonization. Governments can fast-track permitting and interconnection timelines and adopt digital solutions, such as AI, to maximize the use of existing assets, shared open, trusted platforms, and RSS. These elements are essential for AI development in energy, enabling utilities, regulators, and innovators to optimize in real time.  
  1. Governments and private-sector partners must coordinate efforts to accelerate clean energy development and access. Key actions include scaling up technology deployment, streamlining and accelerating permitting processes, developing procurement pathways, and targeted investments to ensure all communities benefit from expanded clean energy.  
  1. Utilize AI as an accelerator of innovation. All tools and technologies empower people globally to build and operate clean, efficient, and resilient energy systems by enabling smarter resource use, improving system efficiency, and fostering innovation in carbon-free energy and conservation. The AI economy has the potential to advance both economic growth and environmental stewardship.  

The IEA estimates that emerging Asian economies must increase annual clean energy investment fivefold to approximately USD 190 billion by 2035 to meet security, climate, and development goals. Grid modernization alone will require $30 billion per year. This highlights the immediacy of coordinated public-private action.  

Microsoft is committed to playing our part as a long-term off-taker of clean energy, a technology partner deploying AI, and a policy collaborator shaping enabling conditions.  

The dual transformation of AI and energy is underway. With the right policies, investments, and partnerships, we can drive emissions reduction and encourage inclusive growth, regional resilience, and technological leadership. The time to act decisively and together is now.  

Source: Powering Progress in Asia: AI and Energy  

Best ChatGPT Alternatives for 2026

Chatbots that use AI are no longer just for testing. In 2026, students, professionals, developers, and businesses in the US all use AI assistants every day. People rely on them to write emails, summarize reports, find information, come up with new ideas, and even fix code. 

A lot of people are looking into other AI assistants, even though ChatGPT is still one of the most popular ones. Some tools are better for coding, privacy, research, or getting things done.  

This guide tells you about the  best ChatGPT alternatives for 2026 and what makes each one unique.  

How Should a Good ChatGPT Alternative Look in 2026?  

ChatGPT can do things that not all AI chatbots can do. Most of the time, the best choices are easy to spot in a few key ways.  

Reasoning and Correctness  

AI tools are now used for important tasks like helping people talk to customers, looking at data, and summarizing documents. That means they have to give correct information and answers that make sense, not just quick ones.  

Connections to tools you use all the time  

When people work with Microsoft Word, Google Docs, Gmail, or Excel, they use AI a lot. AI assistants that work directly with these platforms can help you finish tasks much more quickly.  

Safety and Privacy  

As AI becomes more common in schools and businesses, privacy is becoming more of a concern. Some people prefer platforms that are more honest about how they use their data.  

Skills that are unique  

There are different AI tools for different tasks. Some people are really good at doing research, while others are really good at coding, automating, or writing in a creative way. Most of the best ChatGPT alternatives are really good at one thing.  

The ten best ChatGPT replacements in 2026  

1. Claude (by Anthropic)  

When it was made, AI safety, thoughtful answers, and clear reasoning were all very important. Now, Claude is one of the most trusted options for ChatGPT.  

One of the best things about Claude is that it can handle long papers and hard topics. Many professionals use it to write organized papers, summarize research papers, or look at reports.  

Many people find it easier to work with Claude because he tends to respond in a calm and balanced way.  

Best for  

Writing in long form  

Short versions of research  

Looking at documents  

Advantages  

Answers that are thoughtful and natural  

Does a good job with long texts  

Good at thinking  

Cons  

Not as many integrations as some of its rivals  

2. Gemini from Google  

Bard was the name of Google Gemini before it was changed. It works well with other Google products.  

If you already use Gmail, Google Docs, or Google Drive, Gemini will fit in with your work. It also has good access to real-time web information, which is helpful for research and staying up to date on current events.  

Gemini can also handle more than one mode, so it can work with pictures, text, and other kinds of data.  

Best for  

People who work with Google Workspace  

People who do research and students  

Information that is up to date  

Pros  

Works well with Google  

Access to web data in real time  

Good for studying  

Disadvantages  

Not always the best way to write creatively  

3. Microsoft Copilot  

Microsoft Copilot is now a great AI tool for getting things done. Microsoft 365 and Windows apps come with it built in.  

It can help you write emails in Outlook, make PowerPoint slides, and even write formulas in Excel.  

Copilot can help businesses and professionals who already use Microsoft tools do their daily work much more quickly and easily.  

Best for  

Getting things done at work  

People who use Microsoft 365 for work  

Pros  

Works very well with Microsoft tools  

Useful for everyday tasks  

Strong business features  

Drawbacks  

Works best with other Microsoft software  

4. Perplexity AI  

Instead of a regular chatbot, people often call Perplexity AI an AI-powered search engine.  

It doesn’t just give answers; it also gives sources and citations, which helps users make sure the information is correct. Because of this, it has become well-liked by students, reporters, and analysts.  

If you often need accurate information for reports or research, perplexity can be very helpful.  

Best for  

Looking up facts and doing research  

Homework  

Getting information  

Pros  

Provides sources and citations  

Answers that are quick and right  

Good for studying  

Cons  

Not as focused on creative work  

5. Mistral AI  

Many people are interested in Mistral AI because it has open-weight AI models and is easy for developers to work with.  

A lot of tech teams and startups prefer it because it gives them more freedom and control over how AI systems are used.  

It works especially well for coding, writing technical papers, and supporting more than one language.  

Best for  

Developers, new businesses, and tech projects  

Pros  

Models that are open and adaptable  

Good performance in terms of technology  

It works well with a lot of languages.  

Drawbacks  

Not as simple for beginners  

6. Meta AI  

Meta AI is already built into apps like WhatsApp, Instagram, and Facebook.  

This connection makes it easy for millions of people to reach it. People use Meta AI a lot to get quick answers to questions, come up with ideas for content, or plan what to post on social media.  

Small businesses also use it to get ideas for content and talk to customers online.  

Best for  

Content for social media, quick questions, and everyday use  

Pros  

Easy to get to through Meta apps  

Good for coming up with ideas for content  

Answers that are quick and easy  

Disadvantages  

Some people are worried about their privacy.  

7. Amazon Q  

Amazon Q is mostly for businesses and the cloud, especially those that use Amazon Web Services (AWS).  

It helps DevOps teams and developers find out what’s wrong with cloud systems, learn about AWS services, and keep an eye on infrastructure.  

It focuses on technical things, so it’s not really for casual users.  

Best for  

Cloud-based DevOps engineering teams  

Businesses that use AWS  

Pros  

Works well with AWS  

Useful for fixing problems in the cloud  

Made for use in business  

Drawbacks  

Not made for normal people  

How to Choose the Right AI Tool  

The best AI assistant for you will depend on what you want to do with it.  

Perplexity and other tools can help you find reliable sources if you’re doing research.  

If you need help writing or looking at documents, Claude is a great choice.  

Microsoft Copilot works well with the tools you use at work every day to get things done.  

Developers might prefer Mistral, DeepSeek, or Amazon Q for technical tasks.  

Many people use more than one AI tool at once to get the best results.  

Final Thoughts  

ChatGPT is still one of the best and most adaptable AI assistants available. But in 2026, there will be more choices than ever before.  

Some AI tools are better for coding, being creative, or getting things done than others. Instead of just one AI tool, many professionals now use more than one at a time.  

You can find the AI assistant that works best for you and your work by looking at different platforms and learning what they do best.  

Questions That People Ask a Lot  

1. Is ChatGPT still the best AI chatbot in 2026?  

ChatGPT is still one of the best AI helpers, but there are now better ones for some things, like doing research, coding, or getting things done at work.  

2. What is the best work alternative to ChatGPT?  

Microsoft Copilot and Claude are two of the most popular tools for work.  

3. Which AI chatbot is best for people who care about their privacy?  

People who want better privacy often choose Claude and open-weight AI models like Mistral.  

4. Is Google Gemini a good choice instead of ChatGPT?  

Gemini is great for people who already use Gmail and Google Docs.  

5. Is it okay to use more than one AI chatbot?  

Yes. Many people use more than one tool at once. For instance, they might use Perplexity to learn, Copilot to get things done, and Claude to write.

OpenAI Expands “Operator” Tools – Chat GPT Moves Beyond Chat

OpenAI has introduced Operator, an autonomous AI agent that directly interacts with the web, performing tasks and managing digital workflows. Now available as a research preview for ChatGPT Pro users in the US, Operator can browse the internet, click buttons, complete forms, and make purchases.  

Key Features of Operator 

The operator uses the new computer using agent (CUA) model, which combines GPT for Oz’s ability to process visual information with advanced reasoning taught through practice and feedback.  

  • The operator interacts with any site by viewing pages as screenshots and performing basic actions, eliminating the need for additional API.  
  • An operator automates tasks such as booking travel or ordering groceries online.  
  • For secure actions, the operator lets users handle sensitive steps for safety.  
  • Users can run several tasks in separate conversation threads.  
  • If an issue arises, the operator diagnoses and resolves it to complete the workflow.  

Expansion Beyond Chat 

Operator, together with features such as deep research and model context protocol (MCP), supports AgentTech AI by enabling Skywind’s solution to execute end-to-end actions. The model context protocol (MCP) is a system that extends the AI agent’s ability to manage and use contextual information during complex tasks. Deep research refers to an AI’s ability to automatically gather and summarize relevant information from multiple online sources.  

  • Workflow automation: Operator manages long-running, multi-step tasks, including gathering information, synthesizing it, and executing final actions.  

Access And Availability 

  • Current Status: Operator is available as a research preview for ChatGPT Pro users in the US at operator.chatgpt.com  
  • Future Rollout: OpenAI plans to expand these capabilities to the Plus team and enterprise users.  
  • Limitations: although a capable operator is in early development and may meet challenges on complex interfaces or non-standard websites  

An operator represents an important step toward developing AI agents that serve as personal digital assistants, saving time by automating routine web tasks.  

OpenAI CEO Sam Altman predicts 2025 will be pivotal for AI agents, which automate tasks for users.  

OpenAI is now taking its first substantial step in this area.  

On Thursday, OpenAI launched a research preview of Operator, a general-purpose AI agent that controls a web browser and performs actions independently. US users on ChatGPT’s $200 Pro plan will first access Operator, with future expansion to Plus, Team, and Enterprise users.  

“[Operator] will be in other countries soon,” OpenAI CEO Sam Altman said during a livestream on Thursday. Europe will, unfortunately, take a while.  

The initial research preview is live at operator.chatgpt.com. OpenAI plans to integrate the operator into all ChatGPT clients soon.  

According to OpenAI, operators can automate tasks such as booking, travel, making restaurant reservations, and shopping online. Users can select from categories such as shopping, delivery, dining, and travel, each with its own set of automation options. When users activate Operator, a window appears that displays a dedicated web browser the agent uses to complete tasks and provide explanations of its actions. Users maintain control of their screen while the Operator works in their own browser.  

OpenAI states that the operator uses a computer using an agent model or CUA, which combines the vision capabilities of the GPT-4.0 model with advanced reasoning. The CUA interacts with website front ends. So it does not need developer-facing APIs to access services.  

The CUA clicks buttons, navigates menus, and fills out forms on web pages like a human user.  

OpenAI collaborates with companies such as DoorDash, eBay, Instacart, Priceline, StubHub, and Uber to ensure that operators comply with their terms of service.  

The CUA model is trained to prompt users for confirmation before finalizing tasks with external side effects, such as submitting an order or sending an email. OpenAI writes in materials provided to TechCrunch that it has already proven useful in a variety of cases, and we aim to extend that dependability across a wide range of tasks.  

OpenAI notes that the CUA is not yet reliable across all scenarios.  

OpenAI notes in a support document that operators are currently unable to reliably handle many complex or specialized tasks, such as creating detailed slide shows, operating intake, integrating intricate calendar systems, or interacting with highly customized or non-standard web interface surfaces.  

To ensure safety, OpenAI requires user supervision for certain tasks, such as banking transactions, even if the CUA and operator could perform them independently. For example, users must enter credit card information themselves. OpenAI states that the operator does not collect or take screenshots of any data.  

On particularly sensitive websites, such as email, the operator requires active user supervision. Making sure users can reach, can directly catch, and resolve any potential mistakes. The models might make. OpenAI says in its support materials.  

Although these measures limit the usefulness of the operators, they help prevent unauthorized spending. Google has adopted a similar strategy with its Project Marina AI Agent, which likewise avoids entering sensitive information like credit card numbers.  

Limitations 

Operator has several significant limitations.   

The operator is subject to daily, task-independent rate limits, although it can perform multiple tasks simultaneously. OpenAI notes that dynamic limits are in place. Overall usage limits reset daily.  

Currently, the operator restricts tasks such as sending emails or deleting calendar events for security reasons, even though the CUA is technically capable of performing these actions. OpenAI intends to expand these capabilities, but no timeline has been provided.  

An operator may become unresponsive when facing complex interfaces, password fees, or capture checks. In such situations, it prompts the user to take over according to the open air  

An Agentic Future 

OpenAI has taken a more guarded approach to developing an AI agent than competitors such as Rabbit, Google, and Anthropic, likely due to concerns about the safety risks associated with the technology.  

Allowing AI systems to take actions on the web introduces risks, such as bad actors who could automate phishing scams, DDoS attacks, or perhaps purchase tickets before others. Given ChatGPT’s widespread use, it is essential for OpenAI to implement safeguards against such exploits.  

OpenAI considers the operator sufficiently safe for release in its current form as a research preview.  

The operator employs tools that seek to limit the models’ susceptibility to malicious prompts, concealed instructions, and fishing efforts. OpenAI explains on its website that a monitoring system pauses execution if suspicious activity is detected, with automated and human review. Good pipelines. Continuously update safeguards.  

As part of its broader effort, Operator is OpenAI’s most ambitious attempt at an AI agent.  

Operator introduces features that make ChatGPT more practical similar to Siri or Alexa but demonstrates greater capabilities than previous assistants.  

AI agents are positioned at the next major step in AI with the potential to transform how people use the internet and their computers, rather than only delivering and processing information; agents can, in theory, take actions and complete tasks.  

With the release of OpenAI’s first practical agent, the feasibility of this vision will soon become evident

Source: OpenAI launches Operator, an AI agent that performs tasks autonomously 

Amazon Pushes Kuiper Satellites to Expand US Internet Access

Amazon, previously known as Project Kuiper, is accelerating its $10 billion low Earth orbit (LEO) satellite constellation, now called Amazon Leo, to deliver high-speed internet to unserved and underserved zones. After successful test launches in 2025, Amazon is deploying satellites to compete with SpaceX’s Starlink, with over 100 machines scheduled in the initial network.  

Key Project Kuiper / Amazon Leo Updates 

  • Deployment progress: After successful tests, Amazon has started deploying production satellites to establish a planned constellation of 32,236 satellites.  
  • Competition: the service aims to compete with SpaceX’s Starlink by providing satellite broadband to homes, businesses, and government entities.  
  • Launch partners column: Amazon has secured more than 80 launches with partners such as ULA, ArianeSpace, and Blue Origin, including holding over 20 launches scheduled for 2026.  
  • Deadlines and future: Amazon is working to meet the FCC’s July 2026 deadline to launch at least 1,618 satellites. Prospective extensions are under discussion due to capacity constraints at launch.  
  • Service launch: Although initially expected in 2024, customer service for the network will begin after the first satellite deployments, with the goal of expanding worldwide connectivity.  

Amazon Leo is considered a significant long-term investment to improve internet connectivity through plants, integrating with Amazon Web Services (AWS) for advanced connectivity solutions.  

Seven years ago, Amazon began developing its advanced satellite communications network, now known as Amazon Leo.  

Our vision was clear: billions lack high-speed internet, and many organizations have no reliable connectivity. By deploying satellites in low Earth orbit, we aim to deliver fast, reliable internet to those beyond the reach of current networks.  

We started with a small engineering team and a code name, Project Kuiper, inspired by the Kuiper Belt.  

The Project Kuiper name remained as we secured initial licenses, major launch contracts, and deployed our first production satellites.  

Today, we are launching our permanent brand for the Amazon Leo program. This name references the low Earth orbit satellite constellation powering our network and marks an exciting new chapter for our initiative.  

Our long-term mission remains unchanged, and we are making strong progress. We now operate one of the world’s largest satellite production lines and have developed advanced customer terminals, including the first commercial phased-array antenna supporting gigabit speeds, with over 150 satellites in orbit. Customers and partners such as JetBlue, L3Harris, DirecTV, Latin America Sky, Brazil, and NBN Co., Australia’s national broadband network operator, are already preparing to deploy the service.  

We are continuing to expand our initial satellite constellation and will begin service rollout as we increase network coverage and capacity.  

To learn more or sign up for updates, please visit Amazon Leo.

Source: Project Kuiper is now Amazon Leo 

Tesla Powerwall Demand Surges as US Incentives Drive Installs

Tesla’s energy storage business is experiencing record growth, with over 1,000,000 Powerwall units installed globally as of September 2025. US incentives and the demand for energy independence drove deployments to 31.4 GWh in 2024, adding substantially to Tesla’s profit. The Powerwall 3, featuring an 11.5 kW output and integrated solar inverter, is accelerating adoption, with installations frequently bundled with solar. These trends are supported by sources including Southern Energy Management, Tech Trench, Digitans, and various YouTube analyses.  

Tesla surpassed 1,000,000 global Powerwall installations by September 2025.  

  • In 2024, Tesla set new records by deploying 31.4 GWh of energy storage.  
  • The energy storage segment now accounts for a significant share of Tesla’s profits.  
  • Federal, state, and utility incentives such as the 30% federal tax credit reduced the $13,000 installed cost to $16,500 in 2024, rapidly increasing accessibility.  
  • The latest Powerwall 3, which includes an integrated solar inverter and 11.5 kW capacity, has accelerated adoption since its release, as noted by multiple industry sources in 2025.  

Impact On Market Facts 

  • International orders are surging, with Taiwan’s orders expected to triple by late 2025.  
  • Despite localized battery fires in early 2025, the market has matured, according to industry reports.  
  • The increasing unreliability of the power grid, particularly in Taiwan as of 2025, is driving residential solar and Powerwall installations, according to Industrial Steam Media.  

Tesla’s energy sector, including Powerwall and Megapack, is expanding faster than any other segment of the company, as reported in technology news outlets.  

Tesla’s energy storage business significantly improved an otherwise disappointing earnings report.  

Last year, the company’s profit declined by 45% from 2024, mainly due to lower electric vehicle sales, as investors had expected. Tesla exceeded Wall Street’s profit and sales estimates because of its energy storage business. Is Tesla deployed a record 46.7 gigawatt hours of storage products in 2025, a 48% increase over last year, according to Official Filings.  

The surge was reflected in new financial benchmarks. Mega Pack, Powerwall, and solar installations now provide nearly a quarter of Tesla’s gross profit. Last quarter, Megapack added $1.1 billion to the $3.8 billion annual storage gross profit. Profit, storage, and energy generation revenue grew 26.5% to $12.8 billion.  

These batteries and panels are highly profitable, with 29.8% gross margins nearly twice Tesla’s vehicle margins.  

Given these results, energy storage is expected to play a larger role in the company’s near future.  

Large energy storage projects, such as those installed for utilities or data centers, tend to be milestone-based, with revenue recognized upon achieving specific milestones. In its 10-K filing with the SEC, Tesla said it expects to recognize $4,960,000,000 this year in deferred revenue from projects already underway. That’s more than double what the company recognized in deferred revenue from its storage product projects in 2025.  

The OBBBA eliminated residential storage tax credits, but commercial credits for Megapack and Megablock products last through the mid-2030s, according to Tesla. OBBBA tariffs may raise battery sales prices. Sales rose on higher volume, but Megapaks’ average price fell, indicating greater market competition.  

Yet overall, Tesla remains optimistic about the storage business, even as it navigates these changing market conditions.  

Although these challenges exist, Tesla highlighted in its earnings report that, as AI infrastructure drives rapid load growth, opportunities emerge for energy storage products to stabilize the grid, shift energy when it is needed most, and provide additional power capacity.

Source: Tesla’s energy storage business is growing faster than any other part of the company 

Amazon Expands Box-Free Returns Across US Locations

Key Takeaways 

  • You can now return eligible Amazon items for free at FedEx Office locations nationwide with no label or shipping box required.  
  • Other drop-off locations include Whole Foods Market, The UPS Store, Kohl’s, Staples, and regional partners such as Winn-Dixie, Save Mart, and Goodwill.  
  • Bring your item and QR code to any drop-off location. No packaging or label needed.  

Amazon is making returns simpler than ever.  

Eligible items can be returned for free at more than 10,000 US locations without a box, tape, or label. Most customers are within five miles of a drop-off point, making returns easy.  

More than 1,500 FedEx Office locations have joined Amazon’s network of return drop-off points, along with Whole Foods Market, The UPS Stores store, Kohl’s, and Staples.  

“Convenience doesn’t end once you place your order,” said George Pillai, Vice President of Returns and Re-commerce at Amazon. “We work hard to help customers find products they’ll love and use. But sometimes a return is needed. When that happens, we want the experience to be seamless. We’re always working to make returns simpler; adding FedEx Office locations is another way of making returns fit into customers’ daily lives.”  

Amazon makes returns more convenient by expanding drop-off locations near customers, allowing them to return items as part of their routine without extra packaging or printing.  

Start your return in your Amazon account. Choose a nearby drop-off location and get a QR code. Bring your unpackaged item and its QR code to the location where staff will scan it and prepare it for shipping.  

You can make free returns without a shipping box at three locations:  

FedEx Office 

You may return eligible Amazon items at FedEx Office locations. This allows you to handle returns conveniently while running other errands.  

Whole Foods Market 

Whole Foods Market stores offer free drop-off for eligible Amazon returns, so you can manage your returns while grocery shopping. Many stores have dedicated return counters, while some have self-service kiosks for quick processing.  

The Ups Store 

Many UPS Store locations accept eligible Amazon returns. This option is free unless there’s another free location closer to your delivery address. In rare cases, you may see a $1 fee.  

Kohl’s and Staples Locations 

Kohl’s and Staples stores make it easy to drop off eligible returns while you shop. If there’s a participating location nearby, it will show up as an option when you start a return. At select Save Mart stores in California and Nevada and at Winn-Dixie stores in Florida, return eligible Amazon items during your regular grocery trip.p.  

Goodwill 

If you are in the greater Seattle area, drop off eligible Amazon returns at select Evergreen Goodwill locations. See all return options in the returns center and pick what works best for you. 

Prime members get free delivery on over 300 million items across 35 categories, all covered by Amazon’s A-to-Z Guarantee. Many items are available for free same-day or next-day delivery. 

Source: Amazon expands free returns with no shipping box, tape, or label needed to over 10,000 U.S. locations 

Chrome Adds AI Feature to Boost Speed and Battery Life

Google has released a major update to its Chrome browser to improve how it manages system resources. Starting in late March 2026, Chrome will include new performance detection tools to boost speed and energy efficiency as web pages become more complex. The browser now actively seeks out and reduces the hidden costs of multitasking, rather than relying on fixed power-saving modes. Google now uses an adaptive system that tracks device usage in real time and provides users with helpful tips to keep their devices running smoothly.  

Smarter Performance Detection 

The main new feature is an improved performance detection system running in the background. It checks how each tab affects your computer’s CPU and memory. Previously, users had to find slowdowns themselves, but now Chrome flags tabs that use excessive resources and slow performance. If performance dips, you will get a notification naming the problematic tab and offering a one-click resource fix.  

This new approach is especially helpful for people who often keep many tabs open for work or fun. The system can distinguish between active and idle tabs. Tasks like video calls or file downloads, and idle pages that waste power, are flagged as they occur. Google Chrome has a good feature that helps users keep processes running smoothly without having to dig through complicated menus or task managers.  

Expanding The Memory Saver Framework 

The March 2026 update enhances memory saver mode, giving users finer control over tab deactivation in the browser. Choose which tabs you will need, and which can sleep, freeing memory for other apps. Sleeping tabs reload instantly when reopened, so there is no noticeable delay.  

To clarify, the browser now shows a memory reclaimed value in the performance dashboard. This lets you see exactly how much RAM you have.  

Key benefit: the browser now displays a “memory reclaimed” value in the performance dashboard. This feature shows exactly how much RAM the automated system has saved. There is also a high-performance mode that keeps the selected websites active, ensuring that essential apps like photo editors and financial dashboards perform optimally regardless of the computer’s load.  

Adaptive Energy Management for Longevity 

For laptop and tablet users, a stronger background-throttling system kicks in when your device is unplugged, or the battery is low. Instead of just slowing down animations, it also lowers the frame rate for less important background content and reduces how often inactive tabs refresh.  

These energy‑shaving features are smart enough to adjust to what you are doing. For example, if you are watching a movie in full-screen mode, the browser ensures the video runs smoothly while cutting power to other open tabs. Early tests show that this flexible energy management can boost battery life on a typical Ultrabook by up to 15% during heavy browsing. It does this by reducing spikes, which are small events that keep the processor active, allowing your device to stay in a low‑power state longer.  

Improving The Modern Web Experience 

These updates make browsing more responsive, even on older computers, with software that handles resource management and provides built-in performance insights from Google. Google addresses common web browsing complaints, as efficient browser management is crucial to a good user experience.  

The Invisible Engine of Efficiency 

As browsers improve, using devices becomes smoother and easier. The software now manages battery life and processing power more efficiently, reducing concerns about slowdowns or overheating. Soon, slow computers may become a thing of the past as browsers anticipate our needs and ensure a smooth operation. This allows us to focus on exploring and working while technology quietly maintains performance in the background. 

Source: Expanding Chrome’s AI experiences to India, New Zealand and Canada 

Microsoft Targets Shadow AI Risks With New Security Update

Securing AI is now a fundamental pillar of Microsoft’s modern security. Our AI-first security platform empowers organizations to address today’s threats and safeguard their future.  

A year ago, we launched Microsoft Security Copilot to help defenders quickly detect, investigate, and respond to network incidents. We have now introduced the next step: column AI agents that help with phishing, data security, and identity management. As cyberattacks become more complex and numerous, AI agents are now vital to modern security.  

Phishing attacks remain among the most common and harmful cyber threats. From December to January 2024, Microsoft found over 30 billion phishing emails targeting customers. The volume of these attacks can overwhelm security teams. Teams relying only on manual work and disconnected tools may struggle to quickly sort threats and manage risk.  

The new phishing triage in Microsoft Security Copilot handles routine phishing alerts and attacks, letting human defenders focus on tougher threats and forward-looking security work. This shows how agents can change security.  

Securing and managing AI is still a top priority for organizations. We are excited to bring new features to Microsoft Defender, Microsoft Intra, and Microsoft Purview to help with this.  

Keep reading to discover more about the new agents in Security Copilot and the latest AI security updates, and see how these innovations can support your organization. Reach out to us today and take the next step in strengthening your security with AI.  

Expanding Microsoft Security Copilot With New AI Agent Capabilities 

Microsoft threat intelligence now processes 84,000,000,000,000 signals per day, underscoring how quickly cyberattacks are growing, including 7,000 password attacks per second. To keep up, scaling defenses with AI agents is a must. We’re adding six new security agents from Microsoft and five from our partners to Security Co‑Pilot, which will be available for preview in April 2025.  

Six New AI Agent Solutions From Microsoft Security 

The six new Microsoft Security Copilot agents help teams handle large volumes of security and IT tasks independently, and they work seamlessly with Microsoft security tools. These agents are built for security. Learn from feedback. Adapt to your workflows and follow Microsoft’s zero-trust framework with Teams in control. Agents speed up responses, focus on the biggest risks, and help organizations protect themselves more efficiently.  

Security co-pilot agents will be available throughout Microsoft’s security platform and are designed for the following tasks:  

  • The phishing triage agent in Microsoft Defender automatically classifies phishing alerts, distinguishing between genuine threats and false positives. It provides clear explanations for each decision and refines its detection processes using administrator feedback.  
  • Alert triage agents in Microsoft Purview identify the most important Data Loss Prevention and Insider Risk alerts. They improve their prioritization accuracy over time, using administrator input to refine results.  
  • The Conditional Access Optimization Agent in Microsoft Intra continuously monitors for new users or applications outside current access policies. It flags these gaps, recommends specific policy updates, and offers easy-to-apply fixes for identity teams.  
  • The vulnerability remediation agent in Microsoft Intune ranks vulnerabilities and recommends fixes for speeding up OS patching after admin approval.  
  • The threat intelligence briefing agent in Security Copilot gathers and summarizes the most relevant and timely threat intelligence based on organization-specific attributes and cyber threat exposure levels.  

Security co-pilots and agentic capabilities are examples of how we continue to deliver innovation, leveraging our decades of AI research. See how Agents work.  

Five New Agentic Solutions From Microsoft Security Partners 

Security works best when everyone is involved, and Microsoft is focused on supporting our security community with an open platform. This allows partners to create solutions that benefit customers. Here are five new AI agents from our partners coming to Security Copilot:  

  • The privacy breach response agent from OneTrust examines data breaches and offers tailored guidance to privacy teams on meeting specific regulatory requirements following an incident.  
  • The network supervisor agent from Aviatrix identifies root causes and summarizes issues with the VPN gateway or the site2cloud connection, including outages and failures.  
  • The SecOps tooling Agent from Blue Voyant reviews a Security Operations Center (a team that monitors and responds to security issues) and its controls, then suggests ways to improve security operations, controls, and complaints.  
  • The alert triage agent from Tanium provides analysts with the relevant context for each alert, enabling them to quickly and confidently determine the right response.  
  • The task optimizer agent from Fletch helps organizations predict and prioritize the most important cyber threat alerts, addressing alert fatigue and the challenge of impossible‑to‑prove security.  

New AI-Powered Data Security Investigations and Analysis 

We are also introducing Microsoft Purview Data Security Investigations to help security teams quickly find and address risks related to sensitive data exposure. These investigations use AI-powered content analysis to identify sensitive data and other risks associated with incidents. Investigators can use these understandings to work securely with partner teams and simplify complex tasks, enabling faster mitigation. This solution connects data security investigations to Defender incidents and Purview Insider Risk cases and will be available for preview in April 2025.  

Further Advances in Securing and Governing Generative AI 

A strong cybersecurity foundation drives successful AI transformation. As more organizations adapt generative AI, securing and managing how they create and use AI at work becomes even more important. Our new report, Secure Employee Access in the Age of AI, reveals that 57% of organizations report more security incidents due to AI use, although most recognize the need for AI controls. Sixty percent have not yet implemented them.  

Securing AI is a new challenge, and leaders are especially concerned about data oversharing, new threats and vulnerabilities, and compliance. Microsoft security solutions are designed for AI to help address these issues with new advanced features that protect AI investments, whether for organizations.  

AI Security Posture Management For Multimodal And Multi-Cloud Environments 

Organizations building their own AI solutions need to strengthen security for AI models running on different platforms and clouds. To help with this, Microsoft Defender now offers AI security posture management for Microsoft Azure and Amazon Web Services. It also supports Google Vertex AI and all models in the Azure AI Foundry catalog. Starting in May 2025, this will cover models like Gemini, Gamma, Meta, LLaMA, Mistral, and custom models. With this new multi-cloud support, organizations can see their AI security posture from code to runtime across Azure, AWS, and Google Cloud. Microsoft Defender Health organizations get started with AI security across multiple models and clouds.  

New Detection And Protection From Emerging AI Threats 

AI introduces new risks, such as more revenue for cyberattacks and undiscovered vulnerabilities. The Open Worldwide Application Security Project (OWASP) lists the top risks and solutions for generative AI apps. Starting in May 2025, Microsoft Defender will offer new and improved AI detections for several OWASP-identified risks, such as indirect prompt injection attacks, sensitive data exposure, and wallet abuse. These new detections will help SOC analysts better protect custom AI apps, with added safeguards for Azure OpenAI service and models in the Azure AI Foundry catalog.  

New Controls To Prevent Risky Access And Data Leaks Into Covert AI Apps 

As more people use generative AI (AI that can create text, images, and other content), many organizations are finding that employees are using AI apps that have not been approved by IT or security teams. This unapproved use, known as Shadow AI (the use of AI tools without company oversight), has greatly increased the risk of sensitive data leaks. To help with this, we are announcing the general availability of the NEI web category filter in Microsoft Intranet Internet Access (a service for managing secure internet connections). This feature lets organizations set detailed access permissions and enforce policies about which users and groups can use different AI applications.  

After setting access policies for AI apps, the next step is to stop users from entering sensitive data into them. To help, we are launching a preview of Microsoft Purview Browser Data Loss Prevention (DLP) controls in Microsoft Edge for business. Security teams can now enforce DLP policies and prevent sensitive data from being entered into generative AI apps. This starts with ChatGPT, Copilot Chat, DeepSeek, and Google Gemini. Learn more about our innovations in security for AI.  

New Phishing Protection In Microsoft Teams For Safer Collaboration 

Email is still the main way phishing attacks happen, but collaboration tools are now common targets too. Starting in April 2025, Microsoft Defender for Office 365 will provide built-in protection against phishing and other advanced threats in Teams. Teams will be safer from harmful links and attachments. Thanks to instant scanning, SOC teams will also get full visibility into related attempts and incidents, with alerts and data available in Microsoft Defender.  

Agile Innovation to Build a Safer World 

We continuously enhance Microsoft security by applying the principles of our Secure Future initiative, aiming to deliver strong, comprehensive protection through advanced AI tools. Thank you for joining us in building a safer world. 

Source: Microsoft unveils Microsoft Security Copilot agents and new protections for AI