In-depth articles with more detailed investigations, interesting thought pieces, and more.
Google’s NotebookLM can be fooled by manipulated websites. Developer Ted Benson demonstrated this by presenting his website to Google’s AI crawler with a made-up story about a trip to the moon with a bicycle, balloons, and scuba gear, while human visitors saw the regular page. He warns that this method of feeding LLMs with targeted …
Tech companies are increasingly turning to nuclear energy to power their data centers, driven by rising electricity demands and climate goals. As Justine Calma reports for The Verge, major players like Amazon and Microsoft have secured significant agreements with nuclear power plants, eyeing next-generation small modular reactors still in development. Nuclear energy offers a consistent …
AI-powered tools make legal action cheaper and easier, as the case of the “LexPunk Army” shows. According to the Harvard Business Review, this group of lawyers and developers used an AI bot to submit masses of comments on a proposed law by the U.S. Treasury Department. The flood of 120,000 comments, compared to an average …
In an article for The Atlantic, David Karpf critiques Sam Altman’s optimistic views on AI, particularly following OpenAI’s recent funding announcement, which values the company at $157 billion despite its significant financial losses. Altman’s manifesto, “The Intelligence Age,” predicts transformative advancements through AI, including solving climate issues and achieving superintelligence. However, the author argues that …
An article by Kyle Chayka for The New Yorker discusses the growing concern among social media users about AI using their personal data. Many Instagram users recently shared a message attempting to opt out of AI data usage, but simply posting this message does not change their legal rights. To effectively protect their data, users …
OpenAI (ChatGPT) and Anthropic (Claude) show clear differences in their business models and revenue streams. According to Tanay Jaipuria, OpenAI is expected to generate five times more revenue than Anthropic by the end of 2024. While OpenAI generates 73% of its revenue from ChatGPT subscriptions, 60-75% of Anthropic’s revenue comes from third-party APIs, primarily through …
Google is developing AI with reasoning abilities inspired by the human brain, similar to OpenAI’s o1 model. Several teams at the company are making progress on AI systems capable of solving complex problems in fields such as mathematics and programming. This was reported by Julia Love and Rachel Metz for Bloomberg. Researchers are using a …
Microsoft’s AI assistant Copilot is set for a major upgrade, according to AI chief Mustafa Suleyman. In an interview with WIRED, Suleyman revealed that Copilot will soon feature a human-like voice, visual capabilities, and improved reasoning skills. The goal is to provide users with a trustworthy AI companion that offers emotional support. New features include …
A detailed article in Fortune provides fascinating insights into the turbulent recent months at OpenAI. The releases of AI models GPT-4o and o1 in particular caused internal conflicts between safety and development teams, as reported by Jeremy Kahn, Sharon Goldman, and Kali Hays. CEO Sam Altman pushed for rapid product launches to maintain OpenAI’s leading …
The rapid development of artificial intelligence is creating a huge demand for electricity that is pushing the U.S. power grid to its limits. As Jennifer Hiller reports in the Wall Street Journal, technology companies across the country are looking for power capacity for their AI data centers. But in many places, high-voltage lines are already …