Here at Mind Matters News, a number of us have  described how Internet search engines and AI chatbots can be very wrong or outright deceptive. We’ve also described how human-sourced information on the internet can be incomplete and erroneous or even outright propaganda.

The public is told that AI systems are super smart and have the world’s info at their electronic beck and call. At the same time, it is humans and human organizations who claim professional expertise and so deliver their “truth” via media and Internet. When we wonder about the accuracy of their information, we’re told to “get critical thinking skills.” 

Okay, let’s do that.

On September 5, 2025, Robert F. Kennedy, Jr., Secretary of Health and Human Services, referred to a forthcoming report revealing a potential link between women taking acetaminophen during pregnancy and autism in their children. 

Image Credit: Dragana Gordic - Adobe Stock

Almost immediately, CBS News reached out to various sources for comment and critique of this potential acetaminophen‒autism link. The American College of Obstetricians and Gynecologists (ACOG) gave CBS a statement that day saying:

There is no clear evidence that proves a direct relationship between the prudent use of acetaminophen during pregnancy and fetal developmental issues.

It sounds as if this controversy is pretty much “case closed” because the ACOG experts have refuted the worrisome link. Right?

CBS News thought so. It trumpeted the ACOG statement as a fact-checker’s dream. As critical thinkers, however, let’s examine the statement carefully ourselves.

I should stress before I begin that I am not proposing to enter the medical controversy as such. As a lawyer, I want to discuss what the statements about the evidence do and don’t mean.

“No clear evidence” means evidence exists

Most of us might think the phrase “no clear evidence,” as used by scientists, flows from robust research and reasoning. Not so. The phrase “no clear evidence” does not state an objective fact. It means only that the writer is not persuaded by the evidence that exists.

Litigation lawyers (like me) use this phrase to downplay evidence or plant seeds of doubt about it. The ACOG thus tacitly admitted that there is “evidence” but it chose not to consider it persuasive. That’s all.

That something doesn’t “prove a direct relationship” adds nothing

The ACOG statement next asserts there’s “no clear evidence that proves a direct relationship” — stop right there.  The term “direct relationship” sounds technical but it conceals what might be known. What is a “direct relationship”?  Is it different from an “indirect relationship”?  By denying the “direct relationship,” the statement tacitly admits there is an indirect relationship.

In medicine and epidemiology, the usual kinds of relationships are either causation or correlation.  We find a direct causal relationship when we can explain how Factor A actually causes Outcome B.  On the other hand, we find only a correlation relationship when Outcome B occurs more often than not when Factor A is present, but we don’t know why.

The ACOG statement didn’t mention causation or correlation, so it didn’t say anything meaningful. Saying that there is no clear evidence to “prove” a relationship they never define gives us no information. But it sure sounds authoritative.

“Prudent use” vs. imprudent use?

Next, the ACOG statement narrows its safety claims to address only the “prudent use” of acetaminophen during pregnancy. By this word choice, ACOG left open the leading question: “You’re not denying that imprudent use could cause the harm, right?”

Image Credit: mrallen - Adobe Stock

If ACOG knows there is no link between a mother’s use of acetaminophen and her child’s autism, then it should just say that. But it didn’t.

Is autism just a “fetal development issue”?

Further narrowing their claim, the ACOG statement says it is talking about a potential link between the use of acetaminophen during pregnancy and “fetal developmental issues.” Most of us wouldn’t think twice about the switch  in terminology; surely that’s just a fancy way to refer to autism.

Actually, no.

Fetal development issues include conditions such as congenital heart defects, neural tube defects (spina bifida, anencephaly),  orofacial clefts (cleft lip and/or palate), chromosomal abnormalities (Down syndrome), limb abnormalities (clubfoot, limb reduction defects), and more.

Research says autism has many possible contributing causes. Some of them are genetic and others occur after birth. Autism is not exactly like the fetal development issues listed above. ACOG surely knew this fact but went non-specific by referring to “fetal development issues” instead of focusing on autism.

Beware the negative pregnant statement

As a lawyer, I’ve frequently advised clients about giving statements. I even wrote a book about that. One thing to avoid is the quagmire of “negative pregnant” claims.

A negative pregnant arises when the person denies some part of an allegation, but that denial is “pregnant” with the possibility that the whole allegation is true except the one part that was denied. ACOG’s statement as a whole is a great example of a negative pregnant. The statement reads like an attempt to take a position without committing to anything concrete.

We should not just accept seemingly authoritative claims and answers from AI and human sources in media or on the internet without at least trying to check them out. As in ACOG’s statement, the defects may involve subtle uses of language and their implications.

A presentation at a 2025 Microsoft Research conference described how people tend to think less carefully and critically when they rely on Generative Artificial Intelligence (GenAI) and Internet information engines (my italics):

While GenAI can improve worker efficiency, it can inhibit critical engagement with work and can potentially lead to long-term overreliance on the tool and diminished skill for independent problem-solving. Higher confidence in GenAI’s ability to perform a task is related to less critical thinking effort.

Maybe when we read a broad assertion like the ACOG statement, the first question we should ask is: Does this statement concretely answer the question?  And then: What does this statement leave out by clever word choices?  We ought not assume the AI and authoritative experts’ statements close off all questions. The ACOG statement’s defects, like the embarrassing government agency report in 2024 about widespread drone sightings, should remind us all to ask probing questions rather than accept supposedly authoritative media and AI reports at face value.