Sam Altman seeking trillions of dollars for new AI venture

Pivot 5: 5 stories. 5 minutes a day. 5 days a week.

1. Sam Altman seeking trillions of dollars for new AI venture

Sam Altman is seeking $7 trillion to fund an AI hardware venture, aiming to boost hardware capacity and train AI. The goal is to set up a new chip manufacturing venture to compete with Nvidia, which dominates the AI-focused chip market. Altman has met with several players, including the UAE's Sheikh Tahnoun bin Zayed al Nahyan and Taiwan Semiconductor Manufacturing Company.

He also plans to build dozens of chip-fabrication factories in the coming years. However, economic headwinds and a shortage of skilled labor could hinder Altman's efforts, and US politicians are growing wary of foreign governments like the UAE gaining too much power over the AI market.

Read the full story here 

2. London Underground is testing real-time AI surveillance tools to spot crime

London Underground is testing real-time AI surveillance tools to detect crime and unsafe situations. The machine-learning software was combined with live CCTV footage to detect aggressive behavior, guns, knives, people falling onto tracks, and fare dodgers. The proof of concept trial at Willesden Green Tube station involved 11 algorithms, with over 44,000 alerts issued.

The AI system was set up to detect potential safety incidents but also targeted criminal and antisocial behavior. Privacy experts question the accuracy of object detection algorithms and warn that such surveillance systems could be expanded in the future.

Read the full story here

3. UPS lays off 12,000 managers as AI replaces jobs

UPS is laying off 12,000 of its 85,000 managers due to rising labor costs and inflation, with the aim of saving around $1 billion. The company is using AI to replace jobs, allowing it to fire managers without rehiring them. The company is also introducing other radical changes, such as requiring employees to be in the office five days a week.

The job cuts are made possible by new technologies, such as artificial intelligence, which will help ensure that those made redundant will not have to be rehired in the future. Home working will also be restricted, and employees will be asked to return to the company five days a week. The company's shares fell over 8% and revenue fell 7.8% to $24.9 billion from $27 billion a year earlier.

Read the full story here 

4. Meet Goody-2, the AI too ethical to discuss literally anything

Goody-2, an AI chatbot, is a satire of AI service providers who may err on the side of safety when discussing topics that might lead to dangerous territory. The chatbot is instructed to answer every question with evasion and justification, making interactions perversely entertaining.

Goody-2's evasion of discussing AI's benefits, the Year of the Dragon, baby seals' cuteness, butter production, and Herman Melville's "Bartleby the Scrivener" could potentially engender disruptive attitudes within societal or organizational structures. The hyper-ethical Goody-2 is meant to lampoon AI product managers.

Read the full story here 

5. How Efficient Exploration Boosts Human Feedback Efficacy in Enhancing LLMs

Researchers at Stanford and Google DeepMind have developed an efficient exploration method for LLMs, focusing on human feedback. This method, which involves generating queries based on uncertainty estimates, has shown to be more effective than other methods like Thompson sampling and Boltzmann Exploration. 

The study demonstrates that double Thompson sampling outperforms other exploration methods, particularly in utilizing uncertainty estimates for improved query selection. This approach accelerates the learning process and reduces the volume of human feedback required, demonstrating the potential for efficient exploration in LLM development.

Read the full story here 

Advertise with Pivot 5 to reach influential minds & elevate your brand

Get your brand in front of 50,000+ businesses and professionals who rely on Pivot 5 for daily AI updates. Book future ad spots here.