
AI-generated summary
As generative AI enters the workplace in large numbers, the British government has released draft guidelines for the ethical and sustainable use of AI, aiming to assist civil servants in using artificial intelligence more safely, fairly and sustainably, while paying attention to the environmental costs and resource consumption of AI.
The British government reminds people to make good use of AI and streamline questions to reduce energy consumption. (Drawing by this newspaper)
[Finance Channel/Comprehensive Report] If you usually use AI chatbots such as ChatGPT, will you get used to saying "please" and "thank you" to AI? The British government's latest draft guidelines for the use of AI remind civil servants that there is no need to be too polite when interacting with AI. Unnecessary text such as "thank you" also requires computational processing, which not only increases computing power consumption, but may also bring additional energy and environmental costs. The guide therefore recommends that when using AI, prompt words should be shortened and unnecessary dialogue should be reduced, and lighter AI models should be given priority for simple tasks.
As generative AI enters the workplace in large numbers, even "courtesy" with AI may be re-examined. Tom's Hardware reports that the British government has released a draft guideline called "Using AI ethically and sustainably", hoping to help government employees use artificial intelligence more safely, fairly and sustainably.
Please read on...
One of the most eye-catching suggestions is that users do not need to say "thank you" to AI chatbots.
The reason is not that AI doesn't care about politeness, but that every input and reply requires computing resources. The guidelines point out that in order to improve efficiency, unnecessary words should be avoided when using AI, including polite words that have no actual function, and prompt words should be kept as short and clear as possible.
In addition to "less polite words," the British government also reminded civil servants that not every job requires the use of the most powerful AI model, but that appropriate tools should be selected based on the complexity of the task.
For example, the guide states that when dealing with simpler tasks, you can use Gemini Flash instead of Gemini Pro, which requires higher computing requirements; when using GPT, you can also choose GPT Instant instead of GPT Thinking, which requires more inference operations.
One of the considerations behind this is the environmental cost of AI. The operation of generative AI relies on data centers and a large number of computing devices. The more content the model processes and the higher the computing requirements, the energy required will also increase. Therefore, shortening prompts, avoiding unnecessary back-and-forth conversations, and selecting appropriate models according to the task can all help reduce unnecessary resource consumption.
The guide also reminds that each government department should still abide by its own AI policies, and some units may have other regulations on the AI tools that can be used and how they are used.
It is worth noting that this guideline is still a draft and has not yet been officially finalized. Therefore, rather than saying that the British government "forbids" civil servants from saying thank you to ChatGPT, it is more accurate to say that the draft recommends that civil servants reduce unnecessary AI interactions and develop more efficient and computational resource-saving usage habits.
Grasp the economic pulse with one hand. Click here to subscribe to Free Finance Youtube channel
AI outlook — possibilities, not facts
The British government will officially publish the final version of its guidance on the use of AI in the coming months.
Likely · Within months

During the Mid-Autumn Festival and National Day holiday, when travel is at its peak, bloggers posing as railway staff appeared on the short video platform to teach so-called "backup tips" and guide passengers to use third-party ticket grabbing apps. Many passengers have had their orders mistakenly canceled or suffered multiple deductions because of their trust in the waitlist service of third-party platforms. Railway 12306 officials have refuted the rumors many times, emphasizing that they have not cooperated with any third-party platform. Waiting tickets only come from channels such as refunds and changes, and there is no acceleration channel. Legal experts pointed out that such videos are essentially pseudo-popular science rumors, and bloggers pretending to be official to spread false information may constitute fraud and violate advertising laws. Experts remind passengers to purchase tickets through official channels and treat the waitlist function rationally to avoid being deceived.

Pope Leo stated that concerns about artificial intelligence destroying the world are not 'fake news' and should be taken seriously, in an apparent rebuke of US President Donald Trump, who previously called such fears a 'hoax'. The Pope criticized tech CEOs advocating for no AI guardrails during a press conference on his flight back to Rome after a trip to France.

The US-China AI rivalry creates a strategic dilemma where both nations seek AI's economic and military advantages but face growing challenges in maintaining human control as systems become more autonomous, potentially transforming AI from a state tool into an independent actor in global politics.

The 2026 World Design Capital Conference (WDCC2026) opened in Shanghai on September 28, with the theme of "Design without boundaries, quality and intelligence generated". The conference announced the Frontier Design Innovation Award, and MIT professor Dava Newman won the Outstanding Contribution Award. SAIC uses AI design tools to solve the problem of styling and wind resistance balance. The scale of Shanghai's creative design industry exceeds 1.7 trillion yuan. Xiao Jiale, managing director of Adidas Greater China, said that more than 65% of its products are designed by local teams. The conference lasts until September 30 and includes activities such as the main exhibition, design carnival and City Walk experience season.

The State-owned Assets Supervision and Administration Commission of the State Council of China announced at a press conference that it will increase the opening of artificial intelligence "rejuvenation communities" and promote fair and accessible AI results. The State-owned Assets Supervision and Administration Commission has released 100 AI benchmark scenarios for central enterprises, bringing together more than 30,000 models. In the future, it will focus on promoting the large-scale application of artificial intelligence in industry and other real economic fields.

On September 28, at the 2026 Hangzhou International Talent Exchange and Project Cooperation Conference, the Hangzhou Technology Transfer and Transformation Center officially released version 3.0 of the Wise Man model. The platform is called a "7×24-hour online AI technology manager". It is committed to solving long-standing pain points in the field of scientific and technological achievement transformation by comprehensively upgrading its data base, improving demand analysis penetration and improving supply and demand matching accuracy.