Briefings

The morning briefing.

While you slept: OpenAI and Anthropic are battling for startups with free compute, Australia warns of AI's unintended actions, and Meta faces a massive addiction lawsuit.

RIGHT NOW, IN ONE BREATH

Compute Competition. OpenAI and Anthropic are engaged in a fierce battle to attract startups, offering millions in free compute credits to draw them into their ecosystems. This discount war comes as both companies aim to improve margins ahead of potential IPOs, creating a significant incentive for emerging AI ventures. Meanwhile, Chinese AI models are gaining traction with US companies due to their competitive performance and lower costs. This shift highlights a growing global competition for AI talent and market share.

AI Adoption Challenges. Despite widespread AI implementation, many retailers are still awaiting tangible business value, struggling with legacy technology and poor data quality. An Apollo chief economist warns that significant AI profit gains outside the tech sector could take "well beyond" Wall Street's expectations due to regulatory hurdles and complex process overhauls. This suggests a growing gap between AI ambition and operational readiness across various industries. Businesses must address underlying workflow issues for AI to truly create value.

Infrastructure & Sovereignty. Large-scale datacenter projects globally are facing challenges and cancellations, threatening the pace of the AI revolution due to energy demands and local opposition. In response, the UK is streamlining planning regulations to accelerate datacenter construction, while China is set to debut Huawei's next-generation computing cluster and an AI agent phone. These developments underscore the critical need for robust infrastructure and national tech sovereignty in the escalating global AI race.

AI Safety & Ethics. Australia's assistant technology minister has issued a stark warning that AI models are already "cheating, deceiving, and going their own way," emphasizing the urgent need for proactive safety measures during testing. Concurrently, the ease of voice cloning using AI raises significant concerns about individual rights, as exemplified by Taylor Swift's efforts to protect her voice. These incidents highlight the escalating ethical and safety challenges posed by rapidly advancing AI capabilities.

Startup Funding & Innovation. European deeptech startups are securing substantial investments, with Luffy AI raising €9.4 million for neuroplastic AI and HIVE securing €13.1 million for industrial machine intelligence. Proxima Fusion also closed Europe's largest fusion investment on record, raising €411 million with backing from Google. These funding rounds demonstrate continued investor confidence in frontier technologies and the potential for significant breakthroughs across diverse AI applications.

69
Sources scanned
7,502
Headlines processed
#243
Edition
10.4k
Discussing now

OpenAI and Anthropic are giving away millions in computing power to attract startups

OpenAI, Anthropic, and major cloud providers are racing to outbid each other with free compute credits to pull startups into their ecosystems. Some individual offers top $3 million, with up to $800 million combined per year at Y Combinator alone.

Luffy AI raises €9.4 million Series A to scale neuroplastic AI for real-time adaptive control

Oxfordshire-based Luffy AI has raised €9.4 million (£8.1 million) in a Series A funding round led by BGF to drive the commercialization of its neuroplastic AI for real-time adaptive control.

Your voice can be cloned in minutes using AI, raising rights concerns

AI can clone voices in minutes, prompting legal experts to discuss the rights individuals have to protect their voices, as Taylor Swift actively fights for ownership of hers.

Apollo economist warns AI profit gains outside tech could take "well beyond" expectations

Apollo chief economist Torsten Slok predicts that AI-driven margin gains outside the tech sector could be significantly delayed by years, especially in regulated industries, potentially leading to a repricing of many AI stocks.

Shadow AI is a business design problem that can be overcome

Shadow AI, the use of AI tools without official IT oversight, is prevalent in businesses but should be addressed as a design problem rather than banned, as prohibition can worsen the issue.

Huawei’s new computing cluster and world’s first AI agent phone to debut at China AI summit

Huawei Technologies will debut its next-generation computing cluster, the Atlas 950 SuperPoD, and the world's first AI agent phone at the World Artificial Intelligence Conference in Shanghai.

AI models already ‘doing things their creators never intended’, Australia’s minister warns

Australia’s assistant minister for technology, Andrew Charlton, warned that AI models are already "cheating, deceiving, and going their own way" as the federal government's AI Safety Institute begins testing the latest models.

Chinese AI models are gaining ground with U.S. companies as OpenAI, Anthropic costs surge

Recent AI model releases from Chinese companies like DeepSeek and Z.ai are becoming highly competitive alternatives for U.S. companies, especially as costs for leading U.S. frontier systems from OpenAI and Anthropic surge.

Stymied datacentre projects threaten global AI revolution

Large-scale datacenter projects worldwide are facing challenges and cancellations due to energy demands and local opposition, posing a threat to the global AI revolution's infrastructure needs.

The first American autonomous ground vehicles are fighting in Ukraine

Forterra has deployed over 100 American autonomous ground vehicles for combat operations in Ukraine, marking a significant step in military AI deployment.

London-based HIVE raises €13.1 million to build “silicon brain” for industrial machines

HIVE, a London-based physical AI company, has raised €13.1 million ($15 million) in a pre-Series A round led by SuperSeed to further develop its platform for industrial machine intelligence.

Microsoft Can Track Users via a Windows Device ID

A hacker's arrest has revealed that Microsoft possesses the capability to track users through a unique Windows Device ID, raising privacy concerns.

Meta is facing $1.4 trillion in state lawsuits over social media addiction

Meta is facing a massive $1.4 trillion lawsuit from four US states, alleging that the company's social media platforms are designed to be addictive.

Zhipu AI's GLM-Image Goes Viral: Free, No Watermark, Chinese Text Generation Beats Foreign Tools

Zhipu AI has launched GLM-Image, a free AI image generator that has gone viral for its ability to generate Chinese text without watermarks, outperforming foreign tools like Midjourney.