NerdNewsAugust 01, 2025 |
News & Trends
Quora's Poe releases API for developers
Quora's AI platform Poe releases an API for developers to access various AI models, including text, image, and video generation, with usage tracked via point-based subscription plans.
OpenAI to Launch AI Data Center in Norway
OpenAI is launching its first AI data center in Europe, Stargate Norway, in partnership with Nscale and Aker. The center will have 230 MW capacity, run on 100,000 Nvidia GPUs, and prioritize Norway's AI ecosystem. It will be powered by renewable energy and incorporate closed-loop cooling.
Google AI Maps World
Google's AlphaEarth Foundations model uses AI to map the world in 10-meter squares, providing a comprehensive view of Earth's surface for machines to read. The model is trained on vast amounts of Earth observation data and can efficiently characterize the planet's land and coastal waters. It offers a unified digital representation, or 'embedding,' that computer systems can easily process, making it useful for research into environmental and agricultural concerns.
AI Is Already Disrupting Labor, and Recent Grads Could Be a ‘Lost Generation’
AI is replacing white-collar workers, with executives warning of significant job displacement. Recent graduates may be particularly affected, with some experts warning of a 'lost generation'. While some argue that AI is being used as a scapegoat for economic issues, others believe it is a major factor in the changing labor market.
Meta prepares for gigawatt datacentres to power ‘superintelligence’
Meta is building gigawatt datacentres to power its superintelligence AI models, with a focus on personal AI capabilities and developing the next generation of AI models. The company is investing heavily in its Meta Superintelligence Labs and plans to extend the useful life of its servers to offset increasing infrastructure costs. |
Options & Tutorials
Waymo's Tekedra Mawakana on Autonomous Vehicles
Waymo co-CEO Tekedra Mawakana will discuss the current state of autonomous vehicles and the industry's future at TechCrunch Disrupt 2025. She will cover topics such as rider safety, public trust, regulation, and competitive pressure.
Microsoft Updates Azure AI Speech
Microsoft's Azure AI Speech now uses a zero-shot text-to-speech model called DragonV2.1Neural, allowing for rapid generation of voice replicas with just seconds of sampled speech. The feature supports over 100 languages and can be used for various applications, including chatbot voices and video dubbing. However, the technology also raises concerns about potential misuse, such as creating convincing deepfakes.
Symmetric Cryptography in Practice: A Developer's Guide to Key Management
Symmetric cryptography is widely used due to its compact key sizes, high performance, and quantum resistance. It has three main use cases: ephemeral session keys, self-use keys, and pre-shared keys. Key management is crucial, with best practices including secure key generation, single-purpose keys, aggressive key rotation, and secure storage solutions.
Read This Before You Trust Any AI-Written Code
A recent report found that about half of all AI-generated code contains security flaws, with 45% of tasks completed producing a vulnerability that was part of the Open Worldwide Application Security Project’s top 10 security vulnerabilities. This is a problem as the use of AI in programming is getting more popular, and the surface area for attack is increasing.
Building Reproducible ML Systems with Apache Iceberg and SparkSQL
Apache Iceberg and SparkSQL provide a foundation for building reliable ML systems. Key features include time travel, schema evolution, and ACID transactions, addressing data management challenges in ML infrastructure. This combination improves development velocity, reduces debugging time, and increases confidence in system reliability. |
Launches & Tools
Beijing summons Nvidia over alleged backdoors in China-bound AI chips
Beijing's internet watchdog has summoned Nvidia to discuss alleged backdoors in its H20 AI chips, which are designed for the Chinese market. The move comes after Washington lifted an export ban on the chips, and Nvidia has denied the existence of backdoors. The incident highlights the ongoing tensions between the US and China over semiconductor exports and national security.
UK Regulator Targets Microsoft and AWS with 'Strategic Market Status'
The UK's Competition and Markets Authority (CMA) has proposed designating Microsoft and AWS with 'strategic market status' to address concerns over their dominance in the cloud market. The CMA found that the two companies hold significant unilateral market power, which can harm UK cloud customers. The designation would allow the CMA to introduce targeted measures to promote competition and reduce prices.
Tesla starts sort-of Robotaxi service in San Francisco by invite only
Tesla has started a Robotaxi service in San Francisco, but with a human driver at the wheel. The service is invite-only and is seen as a step towards fully autonomous vehicles. However, the company still faces regulatory hurdles and criticism over its self-driving capabilities.
Ransomware Attacks Escalate to Physical Threats
Ransomware actors are using extreme measures, including physical threats to executives, to pressure victims into paying demands. A new report by Semperis found that 40% of ransomware incidents involved physical threats, with 46% of US-based firms targeted. Threat actors also threatened to file regulatory complaints against victims, with 47% of attacks involving this tactic.
Tesla's Robotaxi Launch in San Francisco
Tesla's newly launched ridehailing service in San Francisco isn't a true robotaxi, as it requires a human driver due to California's regulations. The company has only obtained one of the three necessary permits to operate a commercial robotaxi service in the state. Despite this, Elon Musk plans to expand the service to other states, including Florida, Nevada, and Arizona. |
Quick Links
Your public ChatGPT queries are getting indexed by Google
OpenAI's ChatGPT allows users to share conversations publicly, which can then be indexed by search engines like Google, potentially revealing personal information. This was part of an experiment that has since been ended by OpenAI.
Developer survey shows trust in AI coding tools is falling as usage rises
A survey of 49,000 developers found that while AI tool usage is rising, trust in their accuracy is falling. 45% of respondents reported frustration with 'AI solutions that are almost right, but not quite', leading to debugging issues. Despite this, devs still find AI tools useful for learning and pair programming, but emphasize the need for robust training and critical consideration of AI suggestions.
Microsoft becomes the second $4 trillion company
Microsoft's stock price has risen, passing a $4 trillion market valuation due to better-than-expected earnings and Azure revenue growth. The company's investment in cloud computing positions it well for future AI computing needs.
Microsoft reports massive cloud uptick as CMA questions licensing
Microsoft's latest quarterly results show a 27% growth in its cloud business, with revenue of $46.7bn. The company's Azure cloud service saw a 39% growth, driven by increased adoption of its AI services, including Copilot. However, the UK's Competition and Markets Authority (CMA) has raised concerns over Microsoft's licensing practices, which it believes may be limiting competition in the public cloud market.
Rakuten selects Cisco, Nokia, F5 as partners for 5G SA network
Rakuten Mobile has selected Cisco, Nokia, and F5 as partners to support its 5G Standalone (SA) network in Japan. The partnership aims to simplify operations, reduce costs, and deliver better customer experiences through cloud-native, AI-driven systems. Cisco will provide its mobile Packet Core Portfolio, while Nokia will offer cloud-native network functions and F5 will enhance security and scalability. |
Share NerdNewsShare your affiliate link to get commission!
Thanks for reading,
If you dont want to receive future editions of NerdNews, |