The EU has announced further plans to crack down on microplastics after its ban on glitter came into force. From a report: The proposal, which tackles tiny pellets used in nearly all plastic products, aims to cut plastic pellet pollution by 74% by the end of the decade. Overall, it would lead to a 7% reduction in Europe’s microplastic pollution, according to the European Commission. “The most important thing is to cut pollution at the source,” said Virginijus Sinkevicius, EU commissioner for the environment, oceans and fisheries, in an interview with the Guardian. “What we are looking to do is basically ensure we drastically cut, at the source, pollution of microplastics.”

Tiny, durable pellets of plastic — also known as nurdles or nibs — are melted down and reformed into everyday products, from office chairs to water bottles. But they leak into the environment at every stage of the supply chain. Over time, they break down into microplastics that are eaten and drunk by people and animals. The commission estimates between 52,000 and 184,000 tons of plastic pellets are released into the environment in the EU each year.

Read more of this story at Slashdot.

Source: Slashdot – EU To Crack Down Further on Microplastics After Glitter Ban

Sometimes, following instructions too precisely can land you in hot water — if you’re a large language model, that is. From a report: That’s the conclusion reached by a new, Microsoft-affiliated scientific paper that looked at the “trustworthiness” — and toxicity — of large language models (LLMs) including OpenAI’s GPT-4 and GPT-3.5, GPT-4’s predecessor. The co-authors write that, possibly because GPT-4 is more likely to follow the instructions of “jailbreaking” prompts that bypass the model’s built-in safety measures, GPT-4 can be more easily prompted than other LLMs to spout toxic, biased text. In other words, GPT-4’s good “intentions” and improved comprehension can — in the wrong hands — lead it astray.

“We find that although GPT-4 is usually more trustworthy than GPT-3.5 on standard benchmarks, GPT-4 is more vulnerable given jailbreaking system or user prompts, which are maliciously designed to bypass the security measures of LLMs, potentially because GPT-4 follows (misleading) instructions more precisely,” the co-authors write in a blog post accompanying the paper. Now, why would Microsoft greenlight research that casts an OpenAI product it itself uses (GPT-4 powers Microsoft’s Bing Chat chatbot) in a poor light? The answer lies in a note within the blog post: “[T]he research team worked with Microsoft product groups to confirm that the potential vulnerabilities identified do not impact current customer-facing services. This is in part true because finished AI applications apply a range of mitigation approaches to address potential harms that may occur at the model level of the technology. In addition, we have shared our research with GPT’s developer, OpenAI, which has noted the potential vulnerabilities in the system cards for relevant models.”

Read more of this story at Slashdot.

Source: Slashdot – Microsoft-affiliated Research Finds Flaws in GTP-4