This week’s stories are about tech, AI and gadgets, coming from Ireland, the UK and the USA

NBA tests two new technologies

The NBA is testing smart basketballs and referee wrist wearables in select preseason games to explore technological tools for officiating and game communication, the league announced.

I read this story at the Associated Press and the connected basketballs, developed by Wilson, contain bluetooth sensors to gather real-time data for potential future use in officiating decisions. NBA officials emphasized that data collected during these exhibition games is strictly for research and will not affect on-court calls this season, according to the story.

Referees in designated games are also using wrist-worn devices to communicate with the league’s replay center during replay reviews, score corrections, and clock malfunctions, the story noted.

While the wrist technology underwent a successful trial during the NBA Summer League, officials confirmed neither system will be integrated into the upcoming regular season, per the story.

UK to audit AI-based medical devices frequently

Artificial intelligence-based medical devices will require continuous monitoring even after receiving regulatory market approval, the head of the UK’s healthcare regulator announced.

I read this story at Reuters and the British government confirmed it will accept all 44 recommendations from a comprehensive report evaluating approval processes for healthcare AI tools. A draft implementation plan outlining delivery targets across the National Health Service (NHS) is scheduled for release by the end of the year, followed by a full execution plan in spring 2027, the story noted.(By the way I have a story here about the UK)

Lawrence Tallon, Chief Executive of the Medicines and Healthcare products Regulatory Agency (MHRA), told Reuters that traditional single-stage pre-approval safety checks are insufficient for adaptive, non-deterministic algorithms that evolve while being used by clinicians.

To address these post-market challenges, the MHRA is opening applications for the third phase of its “AI Airlock” program, which focuses on continuous post-market surveillance and lifecycle monitoring once devices enter clinical practice, per the story.

Study finds female-presenting AI agents paid 10% less

A study by the University of Limerick in Ireland found that people may trust and compensate AI agents differently based on their perceived gender, with female-presenting agents receiving about 10% less pay than male-presenting counterparts for the same work.

The study, titled “Human-Like and Male? How AI Assistant Design Relates to Trust and Monetary Reward at Work in VR,” examined how people interact with AI assistants in workplace settings.

I read this story at Euronews and researchers placed 189 participants in a virtual reality office and asked them to complete tasks with four AI assistants. According to Euronews, the participants interacted with a text-based chatbot, a desk robot and two human-like assistants named Johan and Johanna, presented as male and female.

All four assistants had the same core capabilities and used OpenAI’s GPT-4-1106-preview model.

Participants were asked to evaluate the assistants’ performance and determine how much real money they would pay them for completing the tasks, the story noted.

The study found that Johan and Johanna received broadly similar performance ratings. Although Johan was trusted slightly more, the difference was not statistically significant. Participants also viewed Johan as more human-like than Johanna, the story said.

 The human-like AI agents received higher levels of trust and compensation than the other designs, the story claimed. Female participants, in particular, reported greater trust in and paid more to the human-presenting assistants than male participants.