Google

Google pushes AI limits with Ironwood

Google has announced Ironwood, its latest and most advanced AI processor, marking the seventh generation of its custom Tensor Processing Unit (TPU) architecture.

Designed specifically for the growing demands of its Gemini models, particularly those requiring complex simulated reasoning, which Google refers to as ‘thinking’, Ironwood represents a significant leap forward in performance.

Instead of relying solely on software updates, Google is highlighting how hardware like Ironwood plays a central role in boosting AI capabilities, ushering in what it calls the ‘age of inference.’

However, this TPU is not just faster but dramatically more scalable. Ironwood chips will operate in tightly connected clusters of up to 9,216 units, each cooled by liquid and linked through an enhanced Inter-Chip Interconnect.

These chips can also be deployed in smaller 256-chip servers, offering flexibility for cloud developers and researchers.

Instead of offering modest improvements, Ironwood delivers a peak throughput of 4,614 teraflops per chip, alongside 192GB of memory and 7.2 terabits per second of bandwidth, making it vastly superior to its predecessor, Trillium.

Google says this advancement is more than a performance boost, it’s a foundation for building AI agents that can act on a user’s behalf by gathering information and producing outputs proactively.

Rather than functioning as passive tools, AI systems powered by Ironwood are intended to behave more independently, reflecting a growing trend toward what Google calls ‘agentic AI.’

While Google’s comparison to supercomputers like El Capitan may be flawed due to differing hardware standards, there’s no doubt Ironwood is a substantial upgrade. The company claims it is twice as powerful per watt as the v5p TPU, even if the newer Trillium (v6) chip wasn’t included in the comparison.

Regardless, Ironwood is expected to power the next generation of AI breakthroughs, as the company prepares to move beyond its current Gemini 2.5 model.

For more information on these topics, visit diplomacy.edu.

DeepMind blocks staff from joining AI rivals

Google DeepMind is enforcing strict non-compete agreements in the United Kingdom, preventing employees from joining rival AI companies for up to a year. The length of the restriction depends on an employee’s seniority and involvement in key projects.

Some DeepMind staff, including those working on Google’s Gemini AI, are reportedly being paid not to work while their non-competes run. The policy comes as competition for AI talent intensifies worldwide.

Employees have voiced concern that these agreements could stall their careers in a rapidly evolving industry. Some are seeking ways around the restrictions, such as moving to countries with less rigid employment laws.

While DeepMind claims the contracts are standard for sensitive work, critics say they may stifle innovation and mobility. The practice remains legal in the UK, even though similar agreements have been banned in the US.

For more information on these topics, visit diplomacy.edu.

Google blends AI mode with Lens

Google is enhancing its experimental AI Mode by combining the visual power of Google Lens with the conversational intelligence of Gemini, offering users a more dynamic way to search.

Instead of typing queries alone, users can now upload photos or take snapshots with their smartphone to receive more insightful answers.

The new feature moves beyond traditional reverse image search. For instance, you could snap a photo of a mystery kitchen tool and ask, ‘What is this, and how do I use it?’, receiving not only a helpful explanation but links to buy it and even video demonstrations.

Rather than focusing on a single object, AI Mode can interpret entire scenes, offering context-aware suggestions.

Take a photo of a bookshelf, a meal, or even a cluttered drawer, and AI Mode will identify items and describe how they relate to each other. It might suggest recipes using the ingredients shown, help identify a misplaced phone charger, or recommend the order to read your books.

Behind the scenes, the system runs multiple AI agents to analyse each element, providing layered, tailored responses.

Although other platforms like ChatGPT also support image recognition, Google’s strength lies in its decades of search data and visual indexing. Currently, the feature is accessible to Google One AI Premium subscribers or those enrolled in Search Labs via the Google mobile app.

For more information on these topics, visit diplomacy.edu.

Google’s Wiz acquisition propels Israel’s tech sector to new heights

Israel’s high-tech sector surged into 2025 with remarkable momentum, instead of continuing the two-year slump in funding.

According to new figures from Startup Nation Central, private Israeli tech firms raised $3.2 billion through 185 deals in the first quarter, rather than declining as in previous periods. This marked a 12% rise from the previous quarter and a 14% increase year on year.

The most striking figure came from mergers and acquisitions, which reached a record-breaking $35.9 billion across 38 deals.

It was driven by Google’s $32 billion acquisition of cybersecurity firm Wiz—the biggest tech exit in Israeli history and one of the largest globally, instead of a more modest transaction typical of previous years.

Even excluding the Wiz acquisition, the M&A volume still hit $3.9 billion—its highest since the third quarter of 2023. Instead of reflecting a single outlier, the strong figures suggest a broader resurgence in investor confidence and corporate activity.

It signals that global interest in Israel’s innovation sector is gaining strength again, instead of continuing to wane as seen over the past two years.

For more information on these topics, visit diplomacy.edu.

Google report exposes North Korea’s growing cyber presence in blockchain industry

North Korean cyber operatives have expanded their activities by targeting blockchain startups in the United Kingdom and European Union.

A report from Google’s Threat Intelligence Group (GTIG) revealed that IT workers linked to the Democratic People’s Republic of Korea (DPRK) have embedded themselves in crypto projects beyond the United States, across the UK, Germany, Portugal, and Serbia.

These operatives, posing as remote developers, have left compromised data and extortion attempts in their wake.

Affected projects include blockchain marketplaces, AI web applications, and Solana-based smart contracts. Some developers worked under multiple fake identities, using falsified university degrees and residency documents to gain employment.

Payments were routed through services like TransferWise and Payoneer, obscuring funds flowing back to the North Korean regime. Cybersecurity experts warn that companies hiring these workers risk espionage, data theft, and security breaches.

GTIG reports that these cyber operations are generating revenue for North Korea, which has been accused of using overseas IT specialists to finance its sanctioned weapons programmes.

Financial service providers, including Wise, have stated that they monitor transactions closely and report any suspicious activity. With increasing global scrutiny, experts caution businesses to remain vigilant against fraudulent hires in the blockchain sector.

For more information on these topics, visit diplomacy.edu.

Japan targets Apple and Google with new law

The Japan Fair Trade Commission (JFTC) announced on Monday that it has designated Apple Inc., its Japanese subsidiary iTunes K.K., and Google LLC under the new smartphone software competition promotion law.

The law targets dominant IT companies in the smartphone app market, regulating areas like smartphone operating systems, app stores, web browsing software, and search engines.

The primary aim of the law is to prevent these giants from blocking market entry for other companies or giving preferential treatment to their own services. The law will take full effect in December, with the designated companies required to correct any problematic practices.

Apple will be required to allow other companies into the App Store business instead of monopolising it, fostering price competition. Google will be prohibited from displaying its services in search results instead of favouring them.

In response, both companies expressed concerns, with Apple questioning the impact on user experience and Google vowing to engage in discussions to ensure fairness.

For more information on these topics, visit diplomacy.edu.

Gemini AI for kids: A new era of safe, smart learning

Google appears to be working on a child-friendly version of its Gemini AI, offering young users a safer and more controlled experience. A recent teardown of the Google app (version 16.12.39) uncovered strings referencing ‘kid users,’ hinting at an upcoming feature tailored specifically for children.

While Gemini already assists users with creating stories, answering questions, and helping with homework, this kid-friendly version is expected to include stricter content policies and additional safeguards.

Google’s existing safety measures for teens suggest that Gemini for Kids may offer even tighter restrictions and enhanced content moderation.

It remains unclear how Google plans to implement this feature, but it is likely that Gemini for Kids will be automatically enabled for Google accounts registered under a child’s name.

Given global regulations on data collection for minors, Google will reportedly process children’s data in accordance with its privacy policies and the Gemini Apps Privacy Notice.

As AI increasingly integrates into education and daily life, a safer, child-focused version of Gemini could provide a more secure way for kids to engage with technology while ensuring parental peace of mind.

For more information on these topics, visit diplomacy.edu.

Nonprofits receive $10 million boost from Google for AI training

Google.org has announced a $10 million grant initiative aimed at helping nonprofits integrate AI into their operations.

Community foundations in Atlanta, Austin, Columbia, New York City, and San Francisco will distribute the grants, providing nonprofits with tailored AI support to enhance their work.

However, this funding forms part of a broader commitment by Google to improve AI adoption across various sectors.

The initiative includes a generative AI accelerator programme and an AI Opportunity Fund that aims to invest nearly $100 million in AI training and integration programmes for nonprofits.

Over the last year, 20 organisations have benefited from these funds, developing and piloting AI curricula to build practical skills within their communities.

According to Maggie Johnson, Vice President and Global Head of Google.org, recipients report that AI helps them achieve goals in a third of the time and at nearly half the cost.

A six-month-long AI accelerator programme has already provided training to 21 nonprofits, impacting more than 30 million people through AI-powered solutions.

The funding aims to enhance operational efficiency across sectors such as education, health, and workforce readiness.

Organisations like the Tech:NYC Foundation’s Decoded Futures project and Project Evident are leading efforts to promote equitable and responsible AI use, encouraging collaboration between tech leaders and nonprofits.

Nonprofits supported by Google’s funding include global organisations like the World Bank and local initiatives such as Climate Ride and Erika’s Lighthouse.

The funding is expected to drive AI literacy, streamline operations, and enhance the impact of organisations working with limited resources.

Project Evident’s managing director, Sarah Di Troia, emphasised the importance of nonprofits engaging with AI to remain relevant and influential in the evolving technological landscape.

For more information on these topics, visit diplomacy.edu.

Google’s popular search feature gets a rival from Perplexity

AI search company Perplexity is developing a feature similar to Google’s popular Circle to Search, according to CEO Aravind Srinivas. He announced on X that the functionality would be ‘coming soon’ to all Android users, though specific details remain unclear.

A demo video shared by Srinivas showed how users can highlight text in conversations with Perplexity and request further information.

In the demo, a user circled a mention of Roger Federer and asked about his net worth, prompting Perplexity to fetch details from the web. However, since Google has trademarked ‘Circle to Search’, Perplexity may need a different name for its version.

Perplexity has been gaining popularity as an AI-powered search assistant, with some users preferring it over Google’s Gemini. The company recently introduced an AI-driven web browser called Comet, though it remains uncertain whether it will expand beyond smartphones to platforms like Windows and macOS.

For more information on these topics, visit diplomacy.edu.