AI

The OECD Employment Outlook 2023 says that unions have four major concerns about AI: 1. The trustworthy use of  the technology including risks relating to privacy, transparency, security and accountability; 2. The skill demands associated with high re- and up-skilling costs for businesses and workers; 3. The physical and mental health risks e.g. through AI-based surveillance; 4. The risk of ...

Working with AMD, Renesas has created a power supply design, using radiation-tolerant ICs and GaN power devices, for an AMD space-grade ‘Versal adaptive SoC’ – the XQRVC1902, a processor with 400 AI inferencing tiles, dual Arm Cortex-A72 cores, dual Cortex-R5F cores, programmable logic and memory. “ISLVERSALDEMO2Z integrates key radiation-hardened components for power management,” said Renesas. “These ICs are designed to ...

The first public demonstration of Hummingbird, claimed to be the world’s first Optical Network-On-Chip for domain-specific AI workloads, will be at Hot Chips August 27-29 at Stanford University. Hummingbird uses vertically stacked packaging technologies to integrate a photonic chip and an electronic chip into one package serving as the communications network for data centres and other high-performance applications. “Photonics is ...

Wimbledon is using AI for commentaries this year. The AELTC has hooked up with IBM and will use the company’s Watson platform to provide commentary. The service will analyse ball tracking data and give information about the types of shot being used. The AI commentator is available on all Wimbledon’s courts and not just the usual six covered by live human ...

“It’s totally nuts,” says Mustafa Suleyman (pictured), CEO of Inflection AI referring to the huge interest in AI. Suleyman has just raised $1.3 billion with investment coming from Microsoft, Nvidia, Bill Gates and Eric Schmidt the former Google CEO. The total amount it has raised so far is $1.525 billion. In May, Inflection launched a conversational chatbot called Pi. The ...

A guy with impeccable AI credentials has dampened down some of the more over-heated expectations about the technology. Yann LeCun, the French holder of a Turing Prize, said that claims that AI will threaten humans are “preposterously ridiculous“. The idea that we “turn on a super-intelligent system that is going to take over the world within minutes”, said LeCun, “that’s ...

The EU has passed a draft bill on AI regulation which could, after further legislative processes, become law in 2026.  The draft categorises AI as either “unacceptable,” “high,” “limited” or “minimal.”  Unacceptable examples are  AI that manipulates vulnerable groups of people, which leads to predictive policing systems and which employs biometric categorisation using gender, race, ethnicity and religion, among other ...

The UK has hosted the first AUKUS AI and autonomy trial, with the aim of driving artificial intelligence into military use. Service personnel, scientists and engineers from the three AUKUS nations (Australia, United Kingdom, United States) combined for the exercise taking place on the Salisbury Plains. It saw “AI-enabled assets” involved in a collaborative swarm to detect and track military ...

2023 shipments of AI server shipments will grow 38.4% y-o-y, says TrendForce, with a total of 1.2 million AI servers equipped with GPUs, FPGAs and ASICs shipped this year. AI servers constitute nearly 9% of the total server shipments and are expected to be 15% of the market by 2026. Between 2022 and 2026  AI server shipments are expected to ...

In the opening keynote at Google’s  annual developer conference on Wednesday the term ‘AI’ was mentioned 140 times in two hours, says CNET. Alphabet, Meta, Microsoft, Amazon and Apple collectively mentioned the term “AI” almost 200 times in their latest earnings calls, up from less than 40 mentions a year ago. Clearly, Alphabet, Meta and Microsoft are much more involved ...