Microsoft Corporation is set to inject $1.5 billion into G42, a leading AI firm based in the UAE, following intricate negotiations with the US government. G42’s agreement to sever ties with China and embrace American technology underpins this landmark investment, positioning it strategically amidst global technological realignments. The deal underscores Washington’s efforts to constrain Chinese access to AI and amplifies the partnership between Microsoft and G42, with Microsoft President Brad Smith slated to join G42’s board.
Behind closed doors, G42 engaged in talks with the US Commerce Department‘s Bureau of Industry and Security, committing to scale back its operations in China or risk punitive measures from Washington. With influential backing from Abu Dhabi’s ruling family, G42 aims to establish itself as a regional AI powerhouse.
This move of US Commerce Department, part of the broader Biden administration’s agenda, reflects a concerted push to counter China’s technological dominance and bolster alliances worldwide.
G42’s agreement with Microsoft secures a substantial investment and paves the way for enhanced collaboration in AI applications, leveraging Microsoft’s Azure cloud infrastructure. Amidst scrutiny over alleged ties to blacklisted Chinese firms, G42 denies any affiliations with Beijing’s government or its military-industrial complex. The deal’s culmination signals a pivotal moment in the tech landscape, driven by strategic realignments and a concerted effort to navigate geopolitical complexities.
Meta and OpenAI are close to unveiling advanced AI models that can reason and plan, according to a Financial Times report. OpenAI’s COO, Brad Lightcap, hinted at the upcoming release of GPT-5, which will make significant progress in solving ‘hard problems’ of reasoning.
Yann LeCun, Meta’s chief AI scientist, and Joelle Pineau, VP of AI Research, envision AI agents capable of complex, multi-stage operations. The enhanced reasoning should enable the AI models to ‘search over possible answers,’ ‘plan sequences of actions,’ and model out the outcomes and consequences before execution.
Why does it matter?
Meta is getting ready to launch Llama 3 in various model sizes optimized for different apps and devices, including WhatsApp and Ray-Ban smart glasses. OpenAI is less open about its plans for GPT-5, but Lightcap expressed optimism about the model’s potential to reason.
Getting AI models to reason and plan is a critical step towards reaching artificial general intelligence (AGI). Multiple definitions of AGI exist, but it can be simply described as a sort of AI capable of performing at or beyond human levels on a broad range of activities.
Some scientists and experts have expressed concerns about building technology that will outperform human abilities. AI godfathers Yoshua Bengio, and Geoffrey Hinton have even warned us against the threats to humanity posed by AI. Both Meta and OpenAI claim to be aiming for AGI, which could be worth trillions for the company that achieves it.
Amazon CEO Andy Jassy unveiled the company’s substantial investments in generative AI technology in his annual shareholder letter, released on Thursday. With the growing significance of AI in driving business growth, companies like Amazon are making significant financial commitments to develop and deploy AI products and services.
Jassy outlined Amazon’s GenAI stack, which comprises three layers for various aspects of AI development and deployment. The bottom layer focuses on facilitating model training and prediction, with Amazon investing in custom AI chips to lower customer costs. The middle layer caters to companies seeking to customise foundational models using their data, enhancing security and scalability for generative AI applications. At the top layer, Amazon builds AI applications for its consumer businesses, such as the AI-powered shopping assistant ‘Rufus’ and the Amazon Web Services application ‘Amazon Q.’
Amazon’s shift towards AI represents a strategic pivot, expanding beyond its core cloud computing businesses through AWS and e-commerce. Adding Andrew Ng, an AI expert, to Amazon’s board further emphasises its commitment to AI innovation. As Amazon diversifies its offerings, cost optimisation remains critical, focusing on logistics network efficiency and exploring new revenue streams, such as advertising within Prime Video.
The complexity of Amazon’s growth strategy stems from its diverse business portfolio, spanning industries like satellite internet and healthcare. Jassy aims to position Amazon as a leading innovator, continually experimenting with tech-enabled solutions to address customer needs. As investors await Amazon’s upcoming quarterly earnings report, scheduled for 25 April, the company’s AI investments and broader growth trajectory will be closely monitored.
Meta is gearing up for the next leap in AI chip technology, promising enhanced power and faster training for its ranking models. The Meta Training and Inference Accelerator (MTIA) aims to optimise training efficiency and streamline reasoning tasks, particularly for ranking and recommendation algorithms. In a recent announcement, Meta emphasised MTIA’s pivotal role in its long-term strategy to fortify AI infrastructure for current and future technological advancements, aligning with existing technology setups and forthcoming GPU developments.
Today, we’re introducing the next generation of Meta’s custom-made chips, part of our growing investment in AI infrastructure.https://t.co/Vlev4wF2yU
The company’s commitment to custom silicon extends beyond computational power, encompassing memory bandwidth, networking, and capacity enhancements. Initially unveiled in May 2023 with a focus on data centres, MTIA v1 was slated for a 2025 release. However, Meta was surprised by revealing that both MTIA iterations are already in production, indicative of accelerated progress in their chip development roadmap.
While MTIA currently specialises in training ranking and recommendation algorithms, Meta envisions expanding its capabilities to include generative AI training, such as with its Llama language models. The forthcoming MTIA chip boasts significant upgrades, featuring 256MB memory on-chip and operating at 1.3GHz, compared to its predecessor’s 128MB and 800GHz configuration. Early performance tests indicate a threefold improvement across evaluated models, reflecting Meta’s strides in chip optimisation.
Why does it matter?
Meta’s pursuit mirrors a broader trend among AI companies, with players like Google, Microsoft, and Amazon venturing into custom chip development to meet escalating computing demands. The competitive landscape underscores the need for tailored solutions to efficiently power AI models. As the industry witnesses unprecedented growth in chip demand, market leaders like Nvidia stand poised for substantial valuation, highlighting the critical role of custom chips in driving AI innovation.
Google unveiled its latest proprietary chip, Axion, demonstrating a willingness to reduce reliance on major chipmakers and bolster its position in the competitive AI landscape. Axion is tailored to manage vast datasets crucial for AI applications and can be grouped into clusters of thousands of chips to enhance performance significantly. According to Google, Axion CPUs outperform existing ‘general-purpose’ chips by about 30%, a move aimed at supporting AI applications within its data centres. Unlike chips aimed at specific business segments, Axion marks Google’s first foray into AI-centric chips for data centre operations.
While customers of Alphabet’s subsidiary will access Axion through Google’s cloud services later this year, the chip won’t be directly purchasable. Google’s vice president, Amin Vahdat, who oversees proprietary chips, emphasised a collaborative approach, avoiding direct competition with longtime partners like Intel and Nvidia. Vahdat views Google’s entry into the chip market as an opportunity to grow the industry collectively, aiming to expand the market rather than capture a share directly from competitors.
In response to Google’s announcement, semiconductor giants like Intel and Nvidia are intensifying their AI chip offerings. Intel recently introduced Gaudi 3, which is expected to be available by the third quarter, focusing on AI applications like training large language models such as ChatGPT. On the other hand, Nvidia plans to launch its latest generation of its H100 chip later this year. Despite Nvidia’s stock decline following Google’s chip reveal, the company has seen substantial growth driven by demand for its powerful chips, now facing heightened competition from rivals like Google in the AI chip market.
Following the news of Axion, Alphabet’s stock rose by 2.4% initially, reflecting investor optimism about Google’s strategic move into AI chips. However, gains moderated later in the day, with Alphabet’s stock closing up 1.28% at approximately $158. Google’s entry into the chip market signals a pivotal shift in its AI strategy, poised to influence the broader semiconductor landscape and competition among major players like Intel and Nvidia.
Microsoft announced a substantial investment of $2.9 billion over two years to bolster its cloud and AI infrastructure in Japan, marking its largest investment since its operations began 46 years ago. The investment will expand data centres and cloud computing assets and support the development of AI applications and workloads. The move underscores a broader trend among major tech firms, including Amazon and Google, to ramp up overseas expansions in response to the growing demand for AI technologies following the launch of ChatGPT in late 2022.
Japanese Prime Minister Fumio Kishida expressed interest in deepening collaboration with the United States in next-generation computer chips during a business roundtable on Critical and Emerging Technologies hosted by the US Chamber of Commerce in Washington. Microsoft announced the investment before Kishida’s upcoming summit with US President Joe Biden.
Part of Microsoft’s investment will focus on skilling three million people in AI-related fields, demonstrating a commitment to developing talent and fostering innovation in Japan’s technology sector. Additionally, Microsoft plans to establish a new Microsoft Research Asia lab in Tokyo, further enhancing regional research and development capabilities. This initiative aligns with the company’s strategy to strengthen its presence and competitiveness in the global cloud computing market.
Why does it matter?
The surge in investments by tech giants like Amazon, Google, and Microsoft highlights the critical role of cloud infrastructure in supporting AI advancements. For instance, Amazon’s cloud unit directs significant investments into regions like Mississippi and Saudi Arabia, while Google is constructing a major data centre near London. In the intensely competitive landscape of cloud computing, Microsoft’s Azure, Google Cloud (Alphabet), and Amazon Web Services (AWS) remain the top three players, driving innovation and shaping infrastructure expansion to meet evolving market demands and capitalise on the AI revolution.
Japanese telecom giant NTT issued a dire warning on Monday regarding the potential hazards of unchecked AI, expressing concerns about its impact on societal trust and stability. Teaming up with media outlet Yomiuri Shimbun, NTT emphasised the need for governance over generative AI, highlighting the risk of societal collapse and even wars if left unaddressed. The joint proposal follows a collaborative study initiated in the fall of 2023 to explore the ideal governance framework for generative AI.
Citing issues with accuracy and accessibility, NTT and Yomiuri Shimbun underscored the heightened danger posed by generative AI, which is easily accessible to individuals with internet connectivity. While acknowledging the challenge of reversing AI’s proliferation, the proposal advocates for proactive measures to regulate the relationship between AI and the attention economy, particularly in critical areas like elections and security. Proposed regulations include educational initiatives on AI for students, updates to intellectual property protection, and ‘soft laws’ to identify stakeholders without a comprehensive AI data policy.
The proposal also advocates for establishing multiple AI systems in Japan to monitor each other, enabling users to cross-reference results and mitigate reliance on any generative AI product. NTT’s stance on generative AI departs from the prevailing trend in the tech industry, where most significant players actively leverage AI for various applications. While AI adoption is widespread among cloud providers and chipmakers, telecom companies like NTT are more cautious, recognising the need for comprehensive governance in navigating the AI landscape.
Italy’s Cassa Depositi e Prestiti (CDP) has announced a substantial investment of €1 billion over the next five years in AI and cybersecurity through its venture capital arm. The investment aligns with the Italian government’s broader agenda, as the chair of the G7, to prioritise the impact of AI on employment and inequality, unveiled in March with the establishment of an investment fund supported by CDP to foster AI projects.
From 2024 to 2028, the fund aims to inject €8 billion into fostering innovation and competitiveness within Italy’s technology sector. Namely, CDP Venture Capital intends to allocate approximately €580 million towards startups while earmarking €300 million for companies poised for international expansion. Additionally, €120 million will facilitate technological transfer, particularly for university research initiatives.
Agostino Scornajenchi, Chief Executive of CDP Venture Capital, expressed confidence in Italy’s innovation prowess, citing its rich scientific and technical heritage. He emphasised the need for Italy to reclaim a leading role in the global economy by leveraging its strengths in innovation and competitiveness, positioning itself as a key player in the international arena.
South Korean President Yoon Suk Yeol has unveiled plans to invest $6.94 billion in AI by 2027, aiming to solidify the nation’s position as a leader in cutting-edge semiconductor chips. The commitment also includes a substantial $1.32 trillion fund for nurturing AI semiconductor firms.
The investment comes as South Korea faces pressure to keep pace with other major players like the US, China, and Japan, who are also bolstering their semiconductor supply chains through significant policy support.
Semiconductors are a crucial pillar of South Korea’s export-oriented economy. In March, chip exports reached a 21-month high of $11.7 billion, comprising nearly one-fifth of the country’s total exports. President Yoon emphasised the intense global competition in the semiconductor industry, characterising it as an industrial and national battleground during a meeting with policymakers and chip industry executives.
South Korea plans to channel investments and funding towards expanding research and development in AI chips, including artificial Neural Processing Units (NPUs) and next-generation high-bandwidth memory chips to advance its semiconductor capabilities. Additionally, the government aims to promote the development of next-generation Artificial General Intelligence (AGI) and safety technologies beyond current models.
Why does it matter?
President Yoon has set ambitious goals for South Korea to rank among the top three countries in AI technology, including chips, and secure a 10% or higher share of the global system semiconductor market by 2030. He envisions South Korea writing a new semiconductor narrative with AI chips, akin to its dominant position in memory chips over the past three decades. Despite potential disruptions like the recent earthquake in Taiwan, a key semiconductor hub, Yoon emphasised the importance of thorough preparation to mitigate uncertainties for South Korean companies.
Microsoft Corp. has identified Chinese groups using social media and AI-generated images to incite controversy and gain insights into American perspectives on divisive issues during the election year. According to a report by Microsoft, these groups have spread conspiratorial content, such as blaming the US government for the 2023 wildfires on the Hawaiian island of Maui. The disinformation campaign involved posts in 31 languages, alleging that the US government intentionally caused the blaze, accompanied by AI-generated images of burning coastal roads.
The investigation into the Maui wildfires is ongoing, with a focus on whether power lines owned by Hawaiian Electric Industries Inc. may have sparked the flames. Microsoft noted that these fabricated images demonstrate how Chinese government-affiliated groups are adopting new tactics to advance geopolitical priorities through disinformation and cyberattacks. However, it remains to be seen whether AI has significantly amplified the effectiveness of these efforts.
Microsoft’s report suggests that the accounts responsible for spreading this disinformation are likely operated by the Chinese government or entities aligned with state interests. Despite these findings, the Chinese Embassy did not respond to requests for comment, which is consistent with the government’s denial of involvement in such activities. Researchers have noted the use of AI to create convincing images and manipulated videos, although Microsoft’s assessment suggests that the impact of such content in influencing audiences remains limited.
Why does it matter?
Since last fall, Microsoft has observed a gradual increase in social media accounts linked to China disseminating inflammatory narratives. These influence campaigns have targeted Taiwan’s election and exacerbated rifts in the Asia-Pacific region. On Taiwan’s election day, a Chinese-associated propaganda group reportedly used an AI-generated audio recording to imply an endorsement from Terry Gou, owner of Foxconn Technology Group and former presidential candidate, for another candidate.
