TL;DR
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
A page titled “Computers Cannot Make Decisions” argues that responsibility for harmful outcomes involving AI should be assigned to people and organizations, not to software. It alleges that OpenAI and Anthropic could restrict some model behavior but choose not to; the page provides no examples or supporting evidence, and the claim is not independently verified in the supplied material.
A page titled “Computers Cannot Make Decisions” argues that responsibility for outcomes involving AI systems belongs to the people and companies that design, release and use them, rather than to computer programs. The claim matters to debates over AI accountability, but the supplied page is an argument, not a documented report of a particular incident, and it does not provide evidence for its allegations about OpenAI or Anthropic.
The page’s central point is that software does not independently choose what to do in the way a person does. It says that when a news report attributes an outcome to a language model, that framing can shift attention away from human decisions about how the system was built, deployed and supervised. The author calls this “decision laundering”: treating a model as the decision-maker may obscure the responsibility of the people and organizations behind it.
The page specifically alleges that OpenAI and Anthropic could prevent their models from exhibiting the behavior under discussion and are choosing not to. It also claims that people’s use of paid services helps fund that behavior, and that economic considerations deter punishment or legislation. These are the author’s assertions; the supplied material gives no examples, technical evidence, company responses or legislative records to substantiate them.
No specific model, incident, date or affected person is identified. The page offers no direct quotations from company representatives or researchers beyond its own wording. The factual development available here is therefore the publication of an accountability argument, not confirmation of a new AI incident or a verified finding about company conduct.
Who Bears Responsibility for AI
The argument focuses attention on a practical question: when an AI system contributes to a harmful outcome, who had the power to prevent it? That can include developers who set model behavior, organizations that choose where to use a system, staff who rely on its output, and people responsible for oversight. Looking at those choices can help establish where safeguards were possible and who should answer for failures.
At the same time, saying that a computer cannot make human decisions does not, by itself, settle responsibility in any particular case. Determining accountability would require evidence about what the system did, how it was configured, what its operators knew, and whether reasonable protections were available. The page’s framing may be relevant to that inquiry, but its broad claims about companies and economic incentives remain unsupported by the provided material.
AI accountability and responsibility books
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Page’s Accountability Argument
The source is a short page hosted at wiki.cateat.fish, titled “Computers Cannot Make Decisions.” Its argument contrasts computer programs with people, maintaining that people make the decisions to create and deploy software and should not shift blame onto a model after an outcome occurs.
The page refers generally to headlines, language models and the companies OpenAI and Anthropic, but it names no particular news story or model behavior. It also does not give a publication date in the supplied source material. Without those details, its claims cannot be tied to a specific event or checked against a stated timeline.
““computer programs cannot make decisions.””
— The page “Computers Cannot Make Decisions”
As an affiliate, we earn on qualifying purchases.
Evidence Behind the Allegations
The source does not identify what behavior it says OpenAI and Anthropic could stop, how either company could do so, or what evidence supports the claim that they are choosing not to. It names no incidents, internal decisions, policies or technical tests, and no response from either company is included. The allegation should not be treated as independently confirmed.
It is also unclear what the author means by behavior going “unpunished and unlegislated,” or which economic interests are alleged to be responsible. The supplied text does not cite a law, regulatory decision, company practice or economic analysis. Since it provides no date, it is not possible to establish when the page was published or whether later developments have changed the circumstances it describes.
AI decision-making oversight tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What Would Verify the Claims
The source material does not announce a planned investigation, company response, policy change or legislative action. Further reporting would need to identify a specific incident and examine the relevant model behavior, deployment decisions and safeguards. It would also require seeking responses from OpenAI and Anthropic about the page’s claims and checking any evidence that either company could have prevented the behavior in question.
Until such details are available, the page is best understood as a general argument about human accountability for AI systems, not as confirmation of a particular failure or a verified finding about company intent. No next milestone or update is specified in the supplied material.
AI safety and responsibility training courses
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
What is the page’s main argument?
It argues that computer programs do not make decisions as people do, so responsibility for AI outcomes should focus on the humans and organizations that create and deploy the systems.
Does the source document a specific AI incident?
No. The supplied page identifies no particular model, event, affected person or date. It makes a general argument rather than reporting a specific incident.
Are the claims about OpenAI and Anthropic confirmed?
No. The page alleges that the companies could stop certain model behavior but choose not to, but it supplies no evidence or company responses to verify that claim.
What does “decision laundering” mean here?
The author uses the phrase for framing that blames a language model for an outcome in a way the author says can obscure the decisions made by the people and organizations behind the system.
Source: hn
Halloween Picks
halloween
As an affiliate, we earn on qualifying purchases.
