Technology

Google I/O 2025 unveils Ironwood TPU and AI tools that transform virtual communication and automation

Wednesday, 21 May 2025 12:08AM UTC

At Google I/O 2025, the tech giant revealed groundbreaking AI advancements including the Ironwood TPU with tenfold processing power, the immersive video platform Google Beam, intelligent real-time interaction tools, and enhanced AI-powered search and shopping experiences, signalling a new era of personalised, efficient, and accessible technology.

Google I/O 2025 marked a pivotal moment for the tech titan, featuring a range of transformative AI technologies that promise to reshape the user experience across its platforms. Central to the event was Google's commitment to making artificial intelligence not just a tool, but an integral part of everyday applications—from communication to data processing. The keynote was a showcase of the company’s ambitious vision, bringing forward innovations such as new AI models, interactive tools, and experimental features set to roll out over the coming year.

Ironwood TPU: The Engine of AI Advancement

Key to Google’s plans is the introduction of the Ironwood Tensor Processing Unit (TPU), a revolutionary piece of hardware that offers a staggering tenfold increase in performance compared to its predecessor. With a capability of handling 42.5 exaflops per pod, Ironwood is designed to enable businesses to tackle complex AI workloads with unparalleled speed. This immense computational power is expected to redefine the landscape for machine learning and data analytics, opening the door for enterprises to innovate on a large scale. As Google Cloud prepares to make Ironwood available later in 2025, analysts predict that this technology could accelerate the adoption of AI solutions across various sectors, from healthcare to finance.

Google Beam: A New Frontier in Video Communication

Transforming the realm of virtual interactions, Google Beam emerged as an AI-first video communication platform, utilising six cameras to create immersive, 3D light field displays. This ensures not only enhanced clarity but also near-perfect head tracking at 60 frames per second. Integrated into Google Meet, Beam allows for real-time translations—facilitating seamless global collaboration. The ability to breakdown language barriers signifies a considerable advancement, with early adopters slated to access Google Beam devices in late 2025. This tool epitomises how AI can enhance communication, making virtual exchanges feel natural and intuitive.

Gemini Live and Project Astra: Interacting with Intelligence

Among the most exciting features introduced is Gemini Live, developed with assistance from Project Astra. This tool allows users to interact with their environment in real time, identifying objects and dispelling misconceptions, such as mistaking one vehicle for another. Scheduled for a wider rollout on both Android and iOS platforms, Gemini Live aims to deepen engagement between users and their surroundings, while also integrating seamlessly with other Google apps.

The AI-driven capabilities of Gemini will not only enhance personal user experiences but also pave the way for future updates that connect the tool further with Google Calendar, Maps, and tasks, making it quintessential for productivity.

Project Mariner: Automating Tasks Effortlessly

Another transformative innovation is Project Mariner, which empowers AI to operate autonomously on the web. With multitasking capabilities that allow for up to ten simultaneous actions, Mariner can learn from user interactions through a “teach and repeat” feature. Developers can access these functionalities beginning summer 2025, heralding a new era of automation where AI can streamline everyday tasks and enhance productivity significantly.

AI Reinvents Search and Shopping Experiences

The event also unveiled an 'AI Mode' that redefines Google Search into a conversational interface capable of handling more complex queries. This new mode promises to generate expert-level reports instantaneously and is strategically positioned to maintain Google's competitive edge amidst rapidly evolving AI offerings from rivals like OpenAI and Anthropic.

Enhancements to shopping experiences were equally impressive. Google’s revamped 'Try It On' feature now enables users to upload their own images, allowing a more personalised and realistic virtual dressing room experience. Coupling this with the extensive Shopping Graph, Google aims to make online shopping increasingly intuitive and consumer-friendly.

Future Outlook: Subscriptions and Advanced Features

To facilitate the growing demand for AI tools, Google announced two subscription tiers—AI Pro and AI Ultra—broadened access to cutting-edge features at competitive prices. The Pro plan is favoured for regular users, while Ultra targets those seeking exclusive early access to new functionalities. This strategic move not only enhances user engagement but clearly outlines Google’s intent to make artificial intelligence a fundamental component of its ecosystem.

Conclusion: Navigating Towards an AI-Driven Future

Google I/O 2025 showcased the company's bold commitment to an AI-enhanced future, indicating that the next leap in technology is firmly rooted in personalization, efficiency, and accessibility. The innovations presented, from Ironwood's unmatched processing capabilities to the multilayered functionalities of Gemini and Project Mariner, exhibit Google’s overarching strategic vision. As these tools roll out, they promise not only to elevate user experience but also to drive significant industry advancements, reshaping how we engage, collaborate, and create in a highly interconnected world.

Reference Map

Paragraph 1: (1)
Paragraph 2: (1), (2)
Paragraph 3: (1), (3)
Paragraph 4: (1), (4)
Paragraph 5: (1), (5)
Paragraph 6: (1), (2)
Paragraph 7: (1), (6)
Paragraph 8: (1)
Paragraph 9: (1)
Paragraph 10: (1), (4)

Source: Noah Wire Services

More on this

https://www.tgnns.com/news/google-i-o-2025-keynote-with-redefining-ai-innovation/2025/05/20/ - Please view link - unable to able to access data
https://www.axios.com/2025/05/20/google-ai-io-announcements-2025 - At the 2025 I/O developer conference, Google unveiled several new AI-driven features and products, highlighting the company's commitment to integrating artificial intelligence into its core offerings. A key announcement was the introduction of 'Google AI Ultra,' a premium subscription service priced at $250 per month, providing users with early access to the latest AI tools. Google also showcased a prototype of Android XR glasses capable of search, live translation, and photography functions. These developments aim to enhance Google's existing services without disrupting its vital advertising and search businesses. The event places Google in direct competition with other tech giants like Microsoft, which held its Build conference a day earlier, and Anthropic, set to host its first developer conference shortly thereafter.
https://elpais.com/tecnologia/2025-05-20/google-avanza-hacia-la-ia-universal-para-dotar-a-gemini-de-capacidad-de-comprension-de-contexto-planificacion-y-ejecucion.html - During the Google I/O 2025 conference, the company presented significant advancements toward a 'universal AI,' focusing on equipping its Gemini system with capabilities for contextual understanding, planning, and execution. Demis Hassabis and Sundar Pichai highlighted personalization, content generation, and low latency as key elements of progress. Notable innovations include a new search mode with advanced reasoning, the 3D technology Project Starline, and the Google Beam tool for immersive video calls. Additionally, improvements in Gemini Live, AI integration in Google Workspace, AI-assisted shopping, and real-time translation in Google Meet were announced. Gemini will incorporate proactive agents through Project Mariner.
https://www.ft.com/content/b9d13cc7-f0f0-4d0d-9dfd-7d387ac2cdf9 - Google has announced a significant transformation of its search engine, introducing an 'AI mode' that shifts it towards a chatbot-like, conversational experience. Unveiled at its annual I/O developer conference, this move is part of Google's push to compete with AI rivals like OpenAI and Anthropic. 'AI mode' is now available to all U.S. users via Google Search and Chrome, building on last year's 'AI Overviews.' Unlike traditional search, the new mode offers direct AI-generated answers, potentially challenging Google's ad-based revenue model, which generated $50 billion in Q1 2025. CEO Sundar Pichai highlighted the launch as a 'total reimagining of search,' with the goal of making Gemini, Google's large language model, a universal AI assistant capable of performing tasks across devices. The latest Gemini 2.5 version reportedly outperforms rivals in benchmarks, especially in coding. Google also revealed Project Mariner, an AI agent capable of managing web tasks, and previewed Project Astra, a multimodal assistant responding to real-time voice queries via mobile or smart glasses. Subscription tiers for advanced features range from $25 to $250 per month. As privacy advocates raise concerns over data access, Google's integration of AI across its platforms marks a strategic pivot away from its traditional ad-driven model.
https://www.axios.com/2025/05/20/google-ai-shopping-features-virtual-try-on-clothes - At its annual I/O developer conference on May 20, 2025, Google announced significant upgrades to its AI-powered shopping tools. A key innovation is the enhanced 'Try It On' virtual dressing room, which now allows users to upload their own photo to virtually try on clothes, moving beyond previous versions that relied on generic AI models based on selected body types. This personalized experience aims to help consumers better visualize clothing before purchasing. Additionally, Google launched 'AI Mode,' a chat-like search experience that will incorporate shopping features in the near future. These updates leverage Google's expansive Shopping Graph, which includes over 50 billion listings, complete with reviews, prices, and availability from global and local retailers. According to Google's Vice President of Advertising and Commerce, Vidhya Srinivasan, the new tools prioritize personalization by using real user photos instead of stock images or standardized models. Interested users can access the new 'try on' feature via Google Search Labs. These advancements underscore the growing influence of AI in transforming the online shopping experience.
https://www.androidcentral.com/news/live/google-i-o-2025-live-blog - Google I/O 2025 showcased numerous groundbreaking updates across Google's product ecosystem, with a heavy focus on AI and the next phase of Android development. Key announcements included Android 16, which introduces Material 3 Expressive for a more personalized, dynamic UI and improved multitasking features. Wear OS 6 gains similar aesthetic and functional enhancements, maximizing screen usage on smartwatches. Gemini AI was central to the event, now expanding across devices including Android Auto, Google TV, Wear OS, and the upcoming Android XR platform. Gemini Live introduced real-time translation, camera sharing, and screen sharing capabilities, while Gemini 2.5 Flash and Pro models were launched with greater efficiency and multimodal processing. Project Mariner and Agent Mode will let Gemini autonomously complete tasks on a browser, transforming productivity workflows. Veo 3, for AI video generation with native audio, launched alongside deeper Gemini integration in Gmail and Search, including AI Overviews and the new AI Mode. Android XR took the spotlight with previews of upcoming smart glasses from Samsung and Xreal, offering features like real-time navigation, translation, and camera streaming powered by Gemini. Google Beam, a 3D video communication platform, was also unveiled. These developments signal major strides in AI integration and spatial computing across Google's platforms.
https://www.digitaltrends.com/mobile/google-i-o-2025-everything-you-need-to-know/ - Google I/O 2025 is set to be a significant event, with expectations of several major announcements. Android 16 is anticipated to be announced, featuring new functionalities such as notifications with live updates, Bluetooth LE audio sharing via Auracast, and adaptive refresh rates. Google is also expected to introduce Material 3 Expressive, the latest evolution of Material Design 3, which will include new guidelines and 'emotional design patterns' for more engaging transitions and animations. Wear OS 5.1 is expected to receive a release date announcement, bringing new features and improvements to stability. Additionally, Google is likely to provide updates on Gemini AI, including its integration into Google Search, Workspace, and Android, as well as advancements in AI-powered health diagnostics and new hardware products.

Noah Fact Check Pro

The draft above was created using the information available at the time the story first emerged. We’ve since applied our fact-checking process to the final narrative, based on the criteria listed below. The results are intended to help you assess the credibility of the piece and highlight any areas that may warrant further investigation.

Freshness check

Score: 10

Notes: The narrative relates to Google I/O 2025, a recent event held in May 2025, discussing newly announced AI technologies and upcoming product launches scheduled later in 2025, indicating very current content with no signs of recycled or outdated information.

Quotes check

Score: 9

Notes: There are no direct quotations attributed to individuals in the narrative, so no verifiable original quote sources exist. This suggests originality in content rather than repetition of prior quoted sources, warranting a high score.

Source reliability

Score: 7

Notes: The narrative is sourced from tgnns.com, a lesser-known tech news outlet without clear wide recognition or established reputation like BBC or Reuters. This limits certainty regarding reliability, though the content is consistent with typical event coverage.

Plausability check

Score: 9

Notes: The claims about new AI hardware (Ironwood TPU), software tools (Google Beam, Gemini Live, Project Mariner), and service offerings (AI Pro and Ultra subscriptions) align with the current trajectory of Google’s AI strategy and industry trends. However, as some products are announced with expected future rollout dates in 2025, full verification is pending, but nothing appears implausible.

Overall assessment

Verdict (FAIL, OPEN, PASS): PASS

Confidence (LOW, MEDIUM, HIGH): HIGH

Summary: The narrative provides timely information on Google's 2025 AI innovations from a recent keynote event, with original content and no outdated references. While the originating website is not a globally renowned source, the plausibility and freshness of the claims support a high confidence in accuracy and relevance.

Google I/O 2025
Artificial Intelligence
Ironwood TPU
Google Beam
Automation
AI innovation