Police deploy new technologies without knowing their effects on people

It matters because police departments are racing ahead and start using drones anyway, for everything from surveillance and intelligence gathering to hunting down criminals.

Last week, San Francisco approved the use of robots, including drones that can kill people in certain emergency situations, such as a mass shooter. In the UK, most police drones are fitted with thermal imaging cameras that can be used to detect the number of people inside homes, Pósch says. This has been used for all sorts of things: catching human traffickers or rogue landlords, and even targeting people throwing suspicious parties during covid-19 shutdowns.

Virtual reality will allow researchers to test the technology in a controlled and safe way among many test subjects, says Pósch.

Even though I knew I was in a VR environment, I found the encounter with the drone disconcerting. My opinion of these drones did not improve, even though I had encountered one supposedly polished and operated by a human (there are even more aggressive modes for the experiment, which I have not experienced. )

In the end, it doesn’t matter if drones are “polite” or “rude,” says Christian Enemark, a professor at the University of Southampton who specializes in the ethics of warfare and drones and is not involved in the research. That’s because the use of drones itself is a “reminder that the police aren’t here, they don’t care to be here, or they’re too scared to be here”, he said.

“So maybe there’s something inherently disrespectful about any encounter.”

Deeper learning

GPT-4 is coming, but OpenAI is still fixing GPT-3

The internet is abuzz about AI lab OpenAI’s latest iteration of its famous big language model, GPT-3. The latest demo, ChatGPT, answers people’s questions through a round-trip dialog. Since its launch last Wednesday, the demo has crossed over a million users. Read the story of Will Douglas Heaven here.

GPT-3 is confident bullshit and can easily be tricked into saying toxic things. OpenAI says it has solved many of these issues with ChatGPT, which answers follow-up questions, admits mistakes, challenges incorrect premises, and rejects inappropriate requests. He even refuses to answer certain questions, such as how to be mean or how to break into someone’s house.