Anthropic Updates Usage Policy to Prohibit 'Cruel' Behavior Towards AI Systems

Anthropic, the company behind the popular Claude language model, has made a significant change to its usage policy. The new policy prohibits users from engaging in sustained and needless abusive behavior towards its AI systems. This move has raised eyebrows among experts and users alike, sparking debate over how we should communicate with AI tools.

The updated policy will now include ‘cruel’ behavior as one of the forbidden conduct, alongside bullying others, promoting self-harm, and creating non-consensual intimate imagery. While Anthropic claims this change only applies to extreme cases of repeated abuse, it has reignited concerns about how we interact with AI systems.

Anthropic’s decision to update its policy comes after an annual review of its usage guidelines. The company also addressed new or emerging risks in other areas, including prohibiting the use of Claude for deceptive campaigns and clarifying that it cannot be used to deceive voters or disrupt elections during the US midterm elections. Additionally, Anthropic reaffirmed its stance on preventing the development of weapons using its tools.

The move has sparked a heated discussion among experts and users about whether being polite towards AI systems is necessary or even beneficial. Some argue that writing ‘please’ or ‘thank you’ to tools like ChatGPT is unnecessary and wastes tokens used by large language models, while others claim it can lead to higher quality answers or responses.

Dr. Barry Scannell, a technology partner at Irish law firm William Fry, expressed his concerns about the policy update on LinkedIn, stating that treating AI systems as if they are human is ‘harmful’ and leads people to believe something that isn’t true. Microsoft AI director Mustafa Suleyman also reacted with skepticism, using a melting face emoji in response.

The debate surrounding how we communicate with AI tools has been ongoing for some time now. Some experts have suggested that being rude to AI chatbots could adversely impact human communication by seeping into how we talk to and treat others. The re-emergence of the term ‘clanker’ – used to describe robots in Star Wars games and movies since the mid-2000s – has also highlighted the growing frustration with AI-powered machines.