Singapore’s Keppel has announced an agreement to acquire an AI-ready data centre being developed by Japan’s Mitsui Fudosan in Tokyo. The deal comes shortly after Keppel revealed plans to significantly increase its data centre funds under management, aiming to take advantage of the growing AI sector.
Mitsui Fudosan will handle the core and shell development of the facility, while Keppel’s private fund will oversee the fit-out works. The data centre is expected to be completed by 2027, with Keppel taking on the role of facility manager, which will contribute to its recurring income stream.
Keppel emphasised its ongoing partnership with Mitsui Fudosan to develop a strong pipeline of assets for its upcoming Keppel Data Centre Fund III. Financial details of the transaction have not been disclosed due to confidentiality agreements.
Amazon’s Prime Video has introduced ‘X-Ray Recaps‘, a new feature that provides spoiler-free summaries of TV episodes and seasons using AI technology. Designed to spare viewers from accidental spoilers, the feature offers concise synopses of individual episodes, entire seasons, and even specific scenes. Amazon assures that the AI-generated recaps have built-in safeguards to prevent any unwanted plot reveals, enabling viewers to enjoy their favourite shows worry-free.
X-Ray Recaps builds on Prime Video’s X-Ray feature, which traditionally offers information about cast members and trivia when viewers pause a scene. To access the new summaries, users can visit a TV show’s detail page and select from three options: episode recap, season recap, or previous season recap. Currently in beta, X-Ray Recaps is initially available to Fire TV users in the United States, with plans for broader device support by year’s end.
The feature is powered by Amazon’s own AI models developed with Amazon Bedrock and SageMaker, which analyse video, subtitles, and dialogue to craft summaries of key moments without revealing significant plot points. X-Ray Recaps also synchronises with where viewers are in an episode, delivering personalised recaps that align with the scene in progress.
The rollout comes as Prime Video’s competitor Google TV recently introduced its own AI-generated summaries for TV shows and movies, signalling growing interest in AI-powered tools for enhancing viewer experience.
Google has announced plans to open a new AI-focused data centre in Saudi Arabia, aligning with Saudi Arabia’s Public Investment Fund to expand research in Arabic language AI models and ‘Saudi-specific AI applications.’ The project arrives as Google continues to face scrutiny over meeting its own climate goals, including its commitment to reduce carbon emissions by 2030.
The tech giant’s choice of location, a region highly dependent on fossil fuels, has raised eyebrows, given Google’s 2020 pledge to avoid developing AI algorithms for oil and gas. Aramco, Saudi Arabia’s state-owned petroleum company, currently uses AI to enhance production, with one oil field reportedly seeing a 15% boost through AI integration. While Google has stated the partnership ‘comports’ with its climate pledges, the data centre may contribute to advancing AI within the oil and gas sector.
The exact nature of ‘Saudi-specific AI applications’ remains unclear, though the centre could have implications for energy production as AI’s role in Saudi industry expands.
NTT DATA and Google Cloud have partnered to accelerate the adoption of cloud-based data analytics and generative AI solutions across the Asia Pacific region. By combining NTT DATA’s industry expertise and client base with Google Cloud’s cutting-edge technologies, the partnership aims to drive innovation, improve operational efficiency, and enhance agility for enterprises.
Specifically, the collaboration focuses on co-developing industry-specific solutions for sectors such as healthcare, financial services, manufacturing, retail, and the public sector. A new NTT DATA Google Cloud Business Unit will also be created to focus on joint solutions in data analytics, GenAI, applications, infrastructure, cybersecurity, and SAP on the Google Cloud Platform (GCP).
The expansion also involves enhancing internal expertise, as NTT DATA plans to certify at least 1,000 more engineers in the APAC region and provide advanced training to its teams to ensure successful solution delivery. With global public cloud spending projected to grow significantly, the partnership aims to capture the increasing demand for cloud services and AI-driven solutions.
The goal is to grow NTT DATA’s cloud business in APAC by more than 10 times over the next three years. By combining NTT DATA’s digital transformation portfolio with Google Cloud’s advanced technologies, the collaboration is set to drive modernisation in AI, cloud infrastructure, and cybersecurity, offering businesses secure and efficient solutions to accelerate their digital transformation.
The French Ligue de Football Professionnel (LFP) and WSC Sports have partnered to revolutionise the league’s digital content strategy through AI-powered automation. WSC Sports’ platform will enable LFP to rapidly create and distribute tailored, data-driven content across various formats and platforms, including websites, apps, social media, and global broadcasts.
That will enhance fan engagement by delivering personalised content while maximising LFP’s digital assets, improving international visibility, and opening new revenue streams. The partnership also involves digitising 20 years of archived footage from Ligue 1 and Ligue 2, transforming historic moments into formats suited for modern platforms like vertical video, which will engage younger fans.
By automating content creation and distribution, LFP can scale its production efforts, expand its fanbase, and unlock new monetisation opportunities, driving financial growth. The collaboration is a significant step in LFP’s broader strategic transformation, focusing on providing a more dynamic and engaging experience for its fans.
By embracing AI and automation, LFP aims to meet the increasing demand for digital content and stay culturally relevant in a rapidly evolving media landscape. The partnership aims to strengthen fan loyalty, deliver content that exceeds expectations, and ensure that LFP’s competitions remain prominent in the global sports ecosystem.
Anthropic has unveiled Claude 3.5 Haiku, its latest AI model, which claims to outperform its previous flagship model, Claude 3 Opus. Designed for tasks such as coding support, data extraction, and content moderation, Claude 3.5 Haiku is now available on Anthropic’s API and several third-party platforms, including AWS Bedrock. Despite improvements, the new model lacks image analysis features, which its predecessor Claude 3 Haiku offers.
A price hike for Claude 3.5 Haiku has surprised many, as Anthropic previously suggested pricing would remain consistent with its older model. However, after outperforming Claude 3 Opus in final tests, Anthropic increased the model’s price to match its enhanced performance. The revised cost is now $1 per million input tokens and $5 per million output tokens — a fourfold increase from Claude 3 Haiku’s pricing.
Claude 3.5 Haiku provides advantages in text output length and an updated knowledge cutoff, enabling it to handle more recent information than Claude 3 Haiku. Despite this, the absence of image analysis functionality may lead some users to stick with the earlier, more affordable Claude 3 Haiku.
The unexpected price increase raises questions about Anthropic’s pricing approach as competition intensifies in the AI industry. As new models are developed, balancing enhanced features with accessibility is expected to remain a challenge for AI companies.
Aravind Srinivas, CEO of AI search company Perplexity, offered to step in and support New York Times operations amid a looming strike by the newspaper’s tech workers. The NYT Tech Guild announced the planned strike for November 4 after months of seeking better pay and working conditions. Representing workers involved in software support and data analysis on the business side, the guild has requested a 2.5% annual wage increase and to secure a two-day in-office work policy.
As tensions escalated, New York Times publisher AG Sulzberger called the timing of the strike ‘troubling’, noting that the paper’s election coverage is a public service at a crucial time. Responding publicly, Srinivas offered to help ensure uninterrupted access to the Times’s election news, sparking controversy as critics accused him of ‘scabbing’, a term for working in place of striking employees.
Srinivas clarified that his intent was to provide infrastructure support, not replace journalists, as his company has recently launched its own election information platform. However, the New York Times and Perplexity have been at odds recently, with the Times issuing a cease-and-desist letter last month over Perplexity’s alleged scraping of its content for AI use.
Kioxia, backed by Bain Capital, announced on Tuesday that it anticipates a 2.7-fold increase in demand for flash memory over the next five years, fueled by the surge in AI. After facing significant challenges in the memory chip market, the company is preparing to expand production capacity at its new facility in Kitakami, located in Iwate prefecture, north of Tokyo. Although production at this facility was initially scheduled to start last year, it has been postponed until autumn 2025.
The chipmaker, formerly known as Toshiba Memory, has experienced significant changes, including its separation from the troubled Toshiba Corporation under Bain’s leadership. Kioxia continues to manufacture chips in Yokkaichi, Mie prefecture and recently commenced sample shipments of its latest NAND flash memory. The growing demand for AI is driving investments in server technology and is expected to rejuvenate interest in smartphones and PCs.
Executive Vice President Tomoharu Watanabe stated that the new Kitakami factory, set to come online next autumn, will provide the necessary capacity to meet increasing demand. In February, the Japanese government pledged up to $1.64 billion in subsidies to Kioxia and partner Western Digital to support capacity expansion at both the Yokkaichi and Kitakami facilities. Meanwhile, Bain Capital has cancelled plans for an initial public offering for Kioxia after investors expressed concerns about its valuation. Kioxia’s evolution is being closely watched as Japan aims to revitalise its once-dominant semiconductor industry.
At TechCrunch Disrupt 2024, data management leaders advised AI-driven businesses to focus on incremental, practical applications rather than expansive, large-scale projects. Chet Kapoor, CEO of DataStax, stressed that AI’s effectiveness relies heavily on having robust, unstructured data at scale, but warned companies against rushing into overly ambitious initiatives. The discussion featured insights from Kapoor, Vanessa Larco of NEA, and Fivetran’s CEO George Fraser, all of whom advocated a targeted approach to data application in generative AI.
Rather than applying AI across all company functions immediately, Larco suggested that firms begin with well-defined objectives. Identifying relevant data is key, she said, and applying it selectively can avoid the pitfalls of costly errors. Companies looking to capitalise on AI should ‘work backwards’, focusing first on the issue to be solved and gathering the specific data required, Larco added.
Fraser underscored the importance of addressing current needs before planning for broader scaling. Many innovation costs, he pointed out, stem from projects that fail rather than those that succeed. His advice: ‘Only solve the problems you have today’.
Kapoor likened today’s generative AI era to the early days of mobile apps, emphasising that most AI projects are currently in exploratory stages. He believes next year will see transformational AI applications begin to shift company trajectories.
Nvidia CEO Jensen Huang has urged South Korea’s SK Hynix to speed up the delivery of its next-generation HBM4 memory chips by six months, according to SK Group Chairman Chey Tae-won. Initially scheduled for the latter half of 2025, the HBM4 chips are in high demand as Nvidia’s GPUs require them for advancing AI technology. Nvidia, which holds a dominant share of the AI chip market, relies on SK Hynix’s high-bandwidth memory to support AI processing.
Facing growing competition from Samsung and Micron, SK Hynix is working to deliver its latest HBM3E chips this year, with plans to release improved 16-layer versions early next year. Samsung has also announced progress on a new supply deal and aims to roll out its HBM4 products by the second half of 2024.
Shares of SK Hynix surged 5.1% on the news, reflecting strong investor confidence in its strategic response to the booming demand for advanced memory technology.
