Stephen Downes

I said the other day that another word for 'cognitive offloading' might be 'trust'. We don't manually check everything the AI does because we trust it to do (more or less) the right thing. So the title of this post caught my eye; Josh Brake summarizes and to a degree endorses Stephen M. R. Covey's book The Speed of Trust. "Covey decomposes trust into two main elements: character and competence... character, he argues, is composed of integrity and intent... competence can similarly be decomposed into two pieces, capability and results." These all together represent "a strong foundation of character" that ought to be instilled in students, suggests Brake. But is it a good account of trust? I don't think so, because truth is a much broader concept. We can trust things that are not human and do not possess virtue or intent: trust the math, trust the process, trust the ice, trust in the future. Trust isn't a property of the thing being trusted, it is a willingness on our side to grant certain expectations to it regarding the outcome.

This is an introduction to a paper, the full text of which is found here. Michael Feldstein has alo provided some AI structures that both help explain what the paper says and test the predictions offered in the paper. I didn't use the AI components, but I did read the intro post and the paper as a whole, which was well worth the effort. It defies summarization in a short post such as this, but here goes: transformer-based AI (such as ChatGPT) learn complex and apparently rule-based systems (such as language or chess) by preserving distinctions that have predictive import in a given context, and discarding the rest. Feldstein calls this the conservation of predictive meaning (CPM) theory. My assessment is that he is not wrong. I say it that way because I would word things a bit differently and draw slightly different conclusions. What he calls 'distinctions' I would call 'patterns'. What he calls "a general mechanism to reduce predictive surprise" I would call 'salience'. I would not say language learners are "like effective cryptographers", nor would I say they "decode what has been communicated." Overall, though, I think he is on the right track.

I think this is exactly right, and applies to learning technology as well: "In a world where custom code can be created far more easily than it could in the past, communities can more easily build bespoke spaces for themselves. There's no need to adopt a one-size-fits-all platform - even an open source one - when you can ask for the exact features you want." Rather, "What would be needed then are agreed-upon rules about how community platforms behave." This is where it gets tricky, because protocol-writers have historically been over-ambitious in their scope. My view is that syntax belongs to protocols, while semantics belongs to communities. That's (in my view) what Werdmuller describes as "the human stuff that rises to the top when code becomes more of a solved problem."

This is a well-argued and dare I say literate response to critics of screen culture. "When you've closed 800 libraries and gutted the infrastructure through which people build reading communities," writes Doug Belshaw, "blaming screens is a conclusion in search of a cause." Woven through the argument is Belshaw's account of what it means to be literate. "To be literate is to be part of a literate community. This involves sharing references, arguing about ideas, and having the knowledge to participate in discourse." Different communities have different kinds of literacies. It's also a good response, to my mind, to the argument based on 'cognitive offloading' and AI. "These young people weren't less capable than previous cohorts; they were differently capable." Ultimately, "If we want to defend democracy, we should be defending the conditions that make critical engagement with it possible."

Bruce Sterling was cyberpunk back when cyberpunk was a thing, and his voice, though it has mellowed with age, still resonates, now with the silky smooth notes of old oak and maple, seeing things like Jazz and AI for what they are and also what they aren't, reminding us not only that we live and create, but that we fade away and that what carries on is really only those unique notes me make in the ether. I hope to be one of those.

As is so often the case, I have mixed feelings about this discussion. Maia Chankseliani summarizes: "The roadmap links equity with pluralism in a way that goes well beyond the access agenda... genuine inclusion requires going beyond the removal of barriers to entry, to engaging seriously with plural forms of knowing, ways of understanding the world developed across different communities, traditions, and geographies." Sure, we need to do more than just open the doors to traditional universities. And there are different ways of knowing. But we do people a disservice if we open access and they find it's not based in anything like genuine knowledge at all. Telling stories is not to my mind equivalent to a proper scientific enquiry. Opening access also means opening access to the type of learning people desire. And as Chankseliani says, "Precisely how institutions can maintain meaningful quality standards while simultaneously recognising plural epistemologies is a genuine unresolved tension."