Direct answer
The "AI letter to humanity" refers to open letters and public statements in which AI researchers, executives, and public figures warn that advanced artificial intelligence could pose society-scale or existential risk, and call for coordinated safety measures. The best-known example is the March 2023 open letter from the Future of Life Institute, which asked AI labs to pause training systems more powerful than GPT-4 for six months while shared safety protocols were developed [1]. A related, deliberately shorter statement from the Center for AI Safety reduced the same concern to a single sentence: mitigating the risk of extinction from AI should be a global priority alongside pandemics and nuclear war [2]. These documents are advocacy and agenda-setting texts, not regulations, and they carry no legal force.
What Is the AI Letter to Humanity and What Does It Actually Warn About?
The most cited "letter to humanity" is the open letter published in March 2023 under the title "Pause Giant AI Experiments." Its central demand is specific rather than poetic: frontier labs should stop training systems more powerful than GPT-4 for at least six months, and use that window to jointly draft and implement a shared set of safety protocols audited by independent outside experts [1]. The letter frames the request around a race dynamic, arguing that competitive pressure pushes labs to deploy capability faster than safety research can keep up.
What the letter actually warns about is worth separating from how it is often summarized. It does not claim that current chatbots are conscious or malicious; it argues that AI systems with human-competitive capabilities can be applied to propaganda, surveillance, and economic disruption, and that the pace of development is outpacing the institutions meant to govern it [1]. The signature line of the argument is institutional lag, not machine rebellion.
A second, shorter document is frequently folded into the same conversation. The Center for AI Safety's statement on AI risk compresses the argument into one sentence: "Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war" [2]. Its brevity is intentional — a single sentence is easy for a large number of experts to endorse without implying agreement on mechanism, timeline, or remedy.
Read together, the two documents span the range of the "letter to humanity" genre. One is a policy ask with a deadline attached; the other is a consensus statement about priority. Neither specifies a technical standard, and neither binds any organization to anything [1][2].
Who Signed the AI Letter to Humanity, and Why Do Experts Disagree About It?
The signatory lists are the reason these letters travel so far. The pause letter gathered signatures from senior AI researchers, including Turing Award winners, alongside well-known technology executives and public intellectuals, which gave the text unusual visibility outside academic circles [3]. The breadth was the point: a demand for a pause carries weight only if the people asking are the people capable of building the systems.
That breadth also produced the central criticism. Sceptics argued that a voluntary pause is unenforceable, that it would be observed by cautious labs while less cautious actors continued, and that it could entrench the largest incumbents by freezing the competitive landscape in their favour [3]. A six-month moratorium, on this view, is a public-relations gesture rather than a control mechanism.
A second line of disagreement came from inside the signatory pool. Some signatories publicly clarified that they endorsed the spirit of the letter — caution, transparency, shared safety standards — without supporting a literal moratorium on training runs [3]. This gap between signing a text and endorsing every clause is common in open-letter campaigns and is a useful reminder to read the specific wording rather than the headline.
There is also a substantive dispute about the risk model itself. Researchers focused on near-term harms such as bias, misinformation, and labour displacement argue that extinction framing pulls attention and funding away from measurable present-day problems [2][3]. Those focused on frontier capability argue the opposite: that low-probability, high-severity outcomes deserve priority precisely because markets and institutions under-invest in them. Both positions appear in the record, and the letters do not settle the question [3].
How Can Students Make Sure Their Own Writing Is Not Flagged as AI-Generated?
The letters changed how universities talk about AI, and that has a practical consequence: many institutions now run AI-writing detection on submitted work. Turnitin's AI writing report gives an overall percentage along with a segmented breakdown showing which sentences or paragraphs contributed to the score, so the result is inspectable rather than a single opaque number [4]. Understanding that structure matters, because a high overall figure can be driven by a small number of passages.
Detection output is an indicator, not a verdict. Turnitin's own guidance frames the report as a signal that requires human judgement, and acknowledges that formulaic, heavily templated, or heavily edited prose can produce misleading results [4]. That is why well-run integrity processes look at drafting history, reference files, and version control alongside the score rather than relying on the percentage alone.
For students, the reliable defence is process evidence combined with a pre-submission check. Keep dated drafts, notes, and outlines; if you used AI for brainstorming or editing, disclose it according to your institution's policy; and run your own draft through a checker before the deadline so you see what an instructor's system will see [4]. Knowing your score in advance turns a post-submission dispute into a pre-submission fix.
It also helps to write in a way that carries your own reasoning. Paragraphs built from specific sources, named examples, and your own analysis are both better writing and less likely to match the statistical patterns detectors look for [4]. The letters ask humanity to think carefully about AI; the practical version of that for a student is knowing exactly which parts of a draft are yours.
If you want to see the same AI score and similarity report your instructor will see — before the deadline rather than after it — turnitin0 lets you upload a draft and download both PDFs in one checkout. Reports come back in under 15 minutes in 98% of cases, files are checked without being added to Turnitin's student paper database, and more than 100,000 AI and similarity reports have been delivered to 20,000+ students. A single check is $3.80, with prepaid packs bringing it to $2.75 per check.
※ Turnitin0.com - Actual Turnitin AI Report Cover, Score, Flag And Similarity Summary
Get [Real Turnitin](https://www.turnitin0.com/) AI & Similarity Report
FAQ
Is the AI letter to humanity legally binding?
No. It is an open letter — a public advocacy document signed by individuals, not a regulation or contract. It asks labs to adopt a voluntary pause and shared safety protocols [1].
How many people signed the AI letter to humanity?
The pause letter collected signatures from a large cross-section of researchers, executives, and public figures, with the list published and updated publicly [3]. The exact count changed over time as signatures were added and, in some cases, withdrawn.
Why did some signatories distance themselves from it?
Several signatories said they supported caution and shared safety standards but did not endorse a literal six-month moratorium on training runs [3]. Open letters often attract sign-on for the general sentiment rather than every clause.
Does the letter mean AI writing will be detected in my essays?
Not directly — the letters are about frontier model risk, not student assessment. Separately, many universities now run AI-writing detection, and Turnitin's report shows an overall percentage with a sentence-level breakdown that requires human interpretation [4].
What is the difference between the pause letter and the extinction statement?
The pause letter is a specific policy ask with a six-month timeline; the Center for AI Safety statement is a one-sentence claim that AI extinction risk should be a global priority alongside pandemics and nuclear war [1][2].