deceptive

Apps, Gadgets, Media & Entertainment

Deezer Applauds Antitrust Fine Against Apple but Accuses Tech Giant of “Deceptive” DMA Response

Streaming music service Deezer is joining Spotify in cheering the European Union’s €1.84 billion fine imposed on Apple for breaking antitrust rules in the streaming music market. Apple’s new DMA rules, introduced in January, are a complicated means of providing a path forward for app developers to distribute apps from alternative app marketplaces. Deezer is among those developers who drafted an open letter to the EC last week, claiming Apple was making a “mockery” of the DMA. So far, we’ve only heard from MacPaw, the maker of a subscription service for apps, Setapp, which announced it was switching to Apple’s DMA terms last week. However, larger developers, like Apple critics Spotify and Epic Games, as well as tech companies like Meta, Mozilla, and Microsoft have criticized Apple’s new rules.

Dylan Williams
March 4, 2024

Robotics & AI

Anthropologists discover deceptive capabilities of trained AI models

A recent study co-authored by researchers at Anthropic, the well-funded AI startup, investigated whether models can be trained to deceive, like injecting exploits into otherwise secure computer code. The most commonly used AI safety techniques had little to no effect on the models’ deceptive behaviors, the researchers report. Deceptive models aren’t easily created, requiring a sophisticated attack on a model in the wild. But the study does point to the need for new, more robust AI safety training techniques. “Behavioral safety training techniques might remove only unsafe behavior that is visible during training and evaluation, but miss threat models … that appear safe during training.