Reasoning with the majority - chatGPT

#ThinkingWithTheMajority 

#chatGPT has attracted considerable attention since its launch in November 2022, prompting concerns about the quality of its output as well as the potential consequences of widespread use and misuse of this and similar tools.

Virginia Dignum has discovered that it has a fundamental misunderstanding of basic propositional logic. In answer to her question, chatGPT claims that the statement "if the moon is made of cheese then the sun is made of milk" is false, and goes on to argue that "if the premise is false then any implication or conclusion drawn from that premise is also false". In her test, the algorithm persists in what she calls "wrong reasoning".

I can't exactly recall at what point in my education I was introduced to propositional calculus, but I suspect that most people are unfamiliar with it. If Professor Dignum were to ask a hundred people the same question, it is possible that the majority would agree with chatGPT.

In which case, chatGPT counts as what A.A. Milne once classified as a third-rate mind - "thinking with the majority". I have previously placed Google and other Internet services into this category.

Other researchers have tested chatGPT against known logical paradoxes. In one experiment (reported via LinkedIn) it recognizes the Liar Paradox when Epimenides is explicitly mentioned in the question, but apparently not otherwise. No doubt someone will be asking it about the baldness of the present King of France.

One of the concerns expressed about AI-generated text is that it might be used by students to generate coursework assignments. At the present state of the art, although AI-generated text may look plausible it typically lacks coherence and would be unlikely to be awarded a high grade, but it could easily be awarded a pass mark. In any case, I suspect many students produce their essays by following a similar process, grabbing random ideas from the Internet and assembling them into a semi-coherent narrative but not actually doing much real thinking.

There are two issues here for universities and business schools. Firstly whether the use of these services counts as academic dishonesty, similar to using an essay mill, and how this might be detected, given that standard plagiarism detection software won't help much. And secondly whether the possibility of passing a course without demonstrating correct and joined-up reasoning (aka "thinking") represents a systemic failure in the way students are taught and evaluated.

---

See also

Andrew Jack, AI chatbot’s MBA exam pass poses test for business schools (FT, 21 January 2023) HT @mireillemoret

Gary Marcus, AI's Jurassic Park Moment (CACM, 12 December 2022)

Christian Terwiesch, Would Chat GPT3 Get a Wharton MBA? (Wharton White Paper, 17 January 2023)

Related posts: Thinking with the Majority (March 2009), Thinking with the Majority - a New Twist (May 2021), Satanic Essay Mills (October 2021)

Wikipedia: ChatGPT, Entailment, Liar Paradox, Plagiarism, Propositional calculus 

Thinking with the majority - a new twist

I wrote somewhere once that thinking with the majority is an excellent description of Google. Because one of the ways something rises to the top of your search results is that lots of other people have already looked at it, liked or linked to it.

The phrase thinking with the majority comes from a remark by A.A. Milne, the author of Winnie the Pooh.

I wrote somewhere once that the third-rate mind was only happy when it was thinking with the majority, the second-rate mind was only happy when it was thinking with the minority, and the first-rate mind was only happy when it was thinking.

When I wrote about this topic previously, I thought that experienced users of Google and other search engines ought to be aware of how search rankings operated and some of the ways they could be gamed, and to be suitably critical of the fiction functioning as truth yielded by an internet search. And I never imagined that intelligent people would be satisfied with just thinking with the majority. (Although I now suspect that Milne may have been having a dig at his friend G.K. Chesterton.)

The sociologist Francesca Tripodi has been studying how people carry out research on the Internet, especially on politically charged topics. She observes how many people (even those we might expect to know better) are happy to regard search engines as a valid research tool, regarding the most popular webpages as having been verified by the wisdom of crowds. In her 2018 report for Data and Society, Tripodi quotes a journalist (!) explicitly articulating this belief.

I literally type it in Google, and read the first three to five articles that pop up, because those are the ones that are obviously the most clicked and the most read, if they’re at the top of the list, or the most popular news outlets. So, I want to get a good sense of what other people are reading. So, that’s pretty much my go-to.

In other words, thinking with the majority.

However, Professor Tripodi introduces a further twist. She demonstrates that politically slanted search terms produce politically slanted results, and if you go onto your favourite search engine with a politically motivated phrase, you are likely to see results that validate that phrase. She also notes that this phenomenon is not unique to Google, but is shared by all internet search engines including DuckDuckGo.

And this creates opportunities for politically motivated actors to plant phrases (perhaps into so-called data voids) to serve as attractors for those individuals who fondly imagine they are carrying out their own independent research. Tripodi observes a common idea that one should research a topic oneself rather than relying on experts, which she compares with the Protestant ethic of bible study and scriptural inference. And this idea seems particularly popular with those who identify themselves as thinking with the minority (sometimes called red pill thinking).

Zeus' inscrutable decree
Permits the will-to-disagree
To be pandemic.

 

---

 Tripodi explains her findings in the following videos

• Truth and Denial: Searching for Information in the Digital Age (Social Science Matrix @ UC Berkeley, April 2021)
• Reimagine the Internet 2 (Knight First Amendment Institute @ Columbia University, May 2021)
  

Tripodi has also presented evidence to the US Senate Judiciary Committee

• July 16, 2019 – Google and Censorship through Search Engines 
• April 10, 2019 – Technological Censorship and Public Discourse

 

See also  

G.K. Chesterton, Heretics (1905), Orthodoxy (1908)

Joan Donovan, The True Costs of Misinformation - Producing Moral and Technical Order in a Time of Pandemonium (Berkman Klein Center for Internet and Society, January 2020)

Michael Golebiewski and danah boyd, Data Voids: Where Missing Data Can Easily Be Exploited (Data and Society, Updated version October 2019)

Francesca Tripodi, Searching for Alternative Facts: Analyzing Scriptural Inference in Conservative News Practices (Data and Society, May 2018)

Wikipedia: Red pill and blue pill, Wisdom of the crowd 

 

Related posts: You don't have to be smart to search here ... (November 2008), Thinking with the Majority (March 2009)

Thinking with the Majority

A.A. Milne "wrote somewhere once that the third-rate mind was only happy when it was thinking with the majority, the second-rate mind was only happy when it was thinking with the minority, and the first-rate mind was only happy when it was thinking". (War with Honour)

I wrote somewhere once that "thinking with the majority" is an excellent description of Google.

'The suggested improvements (in Google) are just great for those people who want to ask the same questions as everyone else, and get the same answers. Google rankings already depend on the clicks of previous websurfers, and this dependency will become more sophisticated. Google will therefore support, with ever-greater efficiency and effectiveness, an intellectual activity characterized by A.A. Milne (author of Winnie-The-Pooh) as "Thinking with the Majority". '

And as Titus-Armand points out, it is also a good description for

'reliance upon authority in which the “authority figure” is represented by the entire population rather than a single individual or a particular group'.

What about thinking with the minority? There is a popular meme known as "thinking the unthinkable", and I think this is what the third-class mind supposes the second-class mind to be doing.

If you are one of those who are happiest when thinking, please subscribe to this blog and also follow me on Twitter.

---

Update: within a few minutes of this item's being syndicated on Twitter, Al Chou reminds me of a related quote: "A great many people think they are thinking when they are merely rearranging their prejudices." Was it Churchill or William James? The authority of the majority (aka Google) prefers the latter; who am I to argue?

You don't have to be smart to search here ...

... but it helps. 

I'm prompted to write this post by a throwaway remark from David McCoy, in a post on election statistics: "You don’t have to be smart to search nowadays - all you have to do is enter the key snippet."

Ah, but how do you find the key snippet? 

My son had a school essay to write comparing two films, so we thought it would be worth looking on the internet to find some analysis. But if you just search for the names of the films, you just get endless cinema listings and DVD sales, plus a few fairly superficial newspaper cuttings. 

So we tried another tack. Who are the key figures (film theory, media studies, sociology) that might be name-dropped in a serious essay? 

Let's start with Lacan. When we added "Lacan" to the name of one of the films, the search engine suddenly unearthed an entirely different set of web pages, including a bunch of blogs apparently created as part of a high school project (sixth-form) and talking about a set of related films including the two we were interested in. Could we have found these blogs any other way? 

Okay I admit it, my son hasn't read Lacan, hadn't even heard of him, but he had a bit of parental help. The point I'm making here is that sometimes the more knowledge you can put into the search, the more useful the results. 

Even Microsoft sometimes misses important stuff when it searches the internet - for example when checking a brand name. See my post on Google and Longhorn.

Internet search looks rather like a P v NP problem. It's fine for checking unoriginality: for example, if the teacher suspects a student of plagiarism, she can put a suspiciously well-phrased sentence into an internet search engine and confirm that the sentence is not original. It is also fine for finding well-structured material: if you want to check Missouri voting statistics, you can probably find something relevant. (See Missouri loses bellwether status, November 2008.) 

But if you want to find an unusual thought, you will have to find an unusual combination of search terms. You do have to be smart to search here.

 

---

Update March 2023

What I described in this post is now known as prompt engineering. The importance of this has become much more apparent since the emergence of AI tools such as Dall-E and ChatGPT. See James Bridle, The Stupidity of AI (Guardian, 16 March 2023). (The Wikipedia entry on prompt engineering was created in October 2021.)

Meanwhile while there are now automated tools to help teachers detect basic plagiarism, these are not much use against students who try to cheat using AI tools. See my post on Reasoning with the majority - chatGPT (January 2023).

Which develops a view of the limitations of Google I have talked about elsewhere: Thinking with the majority (March 2009)

Trusting Commons

Can we trust Wikipedia? Does Wikipedia give us the truth? James Governor applies the Invisible Hand argument to this question. He points out that there are perhaps just as many errors in traditional encyclopaedias, and mentions both Grove and the Encyclopaedia Britannica.

In our writings, we identify four types of trust. The Encyclopaedia Britannica pretends to offer a form of trust based on authority, but it is in fact a commercial product and relies ultimately on commodity trust. In contrast, Wikipedia offers a form of trust based on the network. Is that good enough?

Instead of asking the binary question: can we trust, yes or no. The key is to ask the discriminating question: for what kinds of question does Wikipedia offer good/better/best answers? Clearly if you want uptodate information about an important but relatively nonpolitical event (such as the Tsunami). the Wikipedia is a pretty good resource. But if you want impartial information about a controversial topic, you may need to exercise caution.

The question of trust/truth in relation to Wikipedia is similar to the question of trust/truth in relation to Google. Can we trust Google? What I'm asking here is not a question about the commercial ethics of the company, but about the use of any Internet search engine as a method of obtaining a true and fair picture of a given topic.

A Google enquiry suffers from some of the same problems as a Wikipedia enquiry. If there is a consensus, you will get the majority view. If there is no consensus, you will get a confused view. You often find the same stupid ideas and ill-informed opinions replicating from one website to another. If there is a coherent minority view, you may never find it. And if you want an original view, you may need to disconnect from the Internet and do some real thinking or discussion instead.

More on Google