Why the UK Government Wants You to Stop Being Polite to AI

If you have found yourself habitually thanking your chatbot after it finishes a task, you are not alone. However, the UK government is officially asking civil servants to cut the pleasantries. As part of a new draft document on using AI efficiently, officials are encouraging staff to ditch the politeness to save on computing power and reduce the overall environmental footprint of digital infrastructure.

The Environmental Cost of AI Politeness

While thanking a machine might seem like a harmless quirk, the government notes that these interactions contribute to unnecessary data processing. Because large language models (LLMs) operate on massive server farms, every word of a prompt consumes energy. The directive outlines specific strategies for using AI efficiently, focusing on sustainability:

• Keep prompts short, clear, and direct.
• Use as few prompts as possible, meaning you do not need to say thank you.
• Only use AI when absolutely necessary for the task at hand.

By streamlining communication with these models, users can reduce the carbon footprint associated with each response. It is a practical shift from treating AI as a conversational partner to treating it as a high-cost utility.

Human Accountability in the Age of Automation

Beyond the environmental concerns, the guidelines serve as a stern reminder of professional responsibility. The UK government emphasizes that while AI can assist in content creation, the human user is ultimately responsible for the final output. There are no excuses for errors, regardless of how complex the software might be:

The guidelines emphasize that you own, and are responsible for, the work produced, leaving no room for civil servants to shift blame onto the AI providers.

This approach highlights a growing divide between the hype surrounding AI and the reality of its implementation in public service. The focus is shifting from simply adopting new technology to managing it with rigorous oversight, accuracy checks, and a keen eye on the actual costs of operation.

Anthropomorphizing LLMs: A Psychological Trap

There is also an implicit psychological argument here. By constantly anthropomorphizing software, users risk over-relying on machines as if they were sentient colleagues. Adopting a more transactional approach—being direct, concise, and professional—helps maintain a healthy boundary between human intelligence and machine-generated data.

As the costs of running AI models continue to climb, we are likely to see more organizations adopt these common-sense efficiency rules. So, the next time you feel the urge to type out a polite gratitude to your AI assistant, remember: the server likely does not care, and the planet would prefer you keep it brief.

Source: Read Original Article