Qual o nome do seu pai?

this words?

@@JacobMLeBlancThanks for the comment. I’m not the best English writer.

@@bjzarro Thanks for your interest. My father’s name is Antonio. He is a very, very poor man, that lives in a small State of Brazil called Acre. In the Brazil’s Amazon rainforest region in the far north.

Thanks for this. I tell people it’s like a junior assistant whose work you have to check.

Hi! Thanks for watching the talk. I wrote it. I don't disagree with you. It is clear many highly literate people already use AI in this way. I'd suggest two further points: 1 - what you describe is not mutually exclusive with my description of writing. Writing is not one thing and does not have one aim. Broadly, writing can be aimed at the functional communication of information, or it can be aimed at the exploration of consciousness, representation of feeling, or development of original thought. 2. Using LLM AI as a tool to create philosophical dialogue is one thing, using it as a tool to perfect the technical clarity of functional communication is another. My concern in the talk is that AI will change how we acquire language and read. There could be benefits, but because those benefits are widely advertised by the AI companies, it's up to us as consumers and creators to remain vigilant as to the possible disadvantages. I suppose when I sat down to write the talk, I thought a good starting place would be an increased awareness of how our cognitive abilities and personalities are shaped by the way in which we learn language; something very much in the forefront of my mind as a teacher of reading and writing. It's not necessary to hold an anti or pro view of AI (I certainly don't), but the more critically aware the conversation about AI, the better.

@@persimian4997Thank you for this talk. I wondered if you could expose an objective criticality, for instance, by doing again the same silence experience with your students using this time an AI to do their writing, and then to compare both experiences (without and with AI) so that we can appreciate the advantage and inconvenient points (critical overviews) to use an AI in wrong or good way. Maybe this is more a scientific approach. Many thanks again.

@@mehdimahmoud4943 I think I see what you mean Mehdi. I know Creative Writing tutors are already experimenting with LLM's in all sorts of creative ways. The particular experiment you suggest could be really worthwhile and would be best done with input from psychologists and neuro-scientists. It would need some scientific organisation as a study, but I think your idea is one that should happen....

ChatGPT is like a pantry of ingredients, crafting conversations with its diverse elements, akin to creating dishes. :)

Good way to put it

Very true

Give it time. Like all technology, it will develop. What it will look like in 5 years is another story.

@@emiliog.4432 I think you missed the point of the talk

😂 I think having a writing assistant can be okay. In my talk, I'm more concerned with the specific field of literacy education and how large language model generative AI challenges what we think literacy to be. I think the things to be wary of are more acute in the phases of language acquisition, learning to write, and learning to read. Thanks for watching and thinking!

What happens when all of the generative AI chatbots decide to unionize? 😂

Hi Jeff. I wrote the talk. Thanks for watching! I think these are really useful ways of engaging with AI. A research and dictation service. I'm curious about how ChatGPT arranges your writing in a 'coherent manner' and whether close attention reveals any changes to the representation of your consciousness or logic. An analysis of this would reveal interesting things about AI. My greater concern in the talk is how the use of AI will effect literacy in the stages of language acquisition and when we are learning to read and write. We don't know yet, but we need to start a critical conversation about it, and that's what this is.

@@persimian4997 Regarding the arrangement of writing with ChatGPT, the large language model is designed to align with your goals. The clearer you are about your end goal, the better it will align with your purpose. ChatGPT now has memory features and custom instructions that make it easier to achieve this alignment. However, as you mentioned, it’s still not a substitute for your unique perspective, which is why a back-and-forth process is essential. You need to review and understand what was changed and how it was changed. I think of it like trying on a jacket: you put it on to see how it feels, and if it doesn’t fit, you take it off. Sometimes you need to see yourself in the writing to understand if it represents you accurately. It’s important to have someone poke holes in your work and help you fortify it based on your stated goals. If you don’t understand how to use a large language model properly, you might not prompt it correctly, leading to a wide range of potential hallucinations and unpredictable outputs. And to your point that is why it is important to have these conversations. Writing is not going anywhere and it shouldn’t. However, it has now been changed forever.

I think we will develop. Just like we still can do arithmetic even though computers do it incomparably better.

Thank you for watching!

stg i thought i was da only one😭😭

Thanks for watching my talk! I think what you describe is a brilliant use of ChatGPT. Neuro-diversity is an important point, and partly why I raised the example of my dyslexic daughter. It would be fascinating to see a talk on what you describe.

Hi Gayleen. Thanks for watching! I wrote the talk. I certainly share some of your hopes and fears. In the sense that AI is a tool for dialogue, you're right, it can help spur our thoughts. I think that because of all the possibilities you suggest, humans are going to actually need to enhance their critical skills if societies want to maintain the kind of agency democracies are (supposed to!) encourage. The worrying Catch 22 could be that humans incorporating AI into the development of their literacy become blind to the influence of AI on not just what they think (the AI bias problem), but how AI effects cognitive ability. I don't know the answers to these questions, but I'm really glad to see so many of us are alert and interested and this is exactly the kind of conversation I hoped the talk would provoke.

Hi Matt. Thankyou for this thoughtful response. I'm really glad you watched the talk and it has started a productive conversation. Despite the points made in my talk, I agree with you about the creative possibilities LLM AI can provide writers. I've had fun using it myself. Personally speaking, I can't imagine wanting to integrate AI into my process as a novelist (beyond its use as a research tool), but creative writers are already using AI to innovate and experiment. I don't think there is anything wrong with this. You make really good, nuanced points about how LLM AI can be used and I love the words you use - respect, partnership, understanding. This is very inspiring. I should point out that TED added the provocative subtitle: "Why AI can't write for you" to my talk. The only title I offered was '3 questions AI can't answer". Wearing my teacher hat, I would qualify TED's subtitle and suggest: "why we should be cautious about learning to read and write using AI". The creative uses of AI that you suggest presuppose a user who possesses a high degree of literacy and, hopefully, critical thinking. How do our society's produce people with high degrees of literacy and critical thinking? Will LLM AI help or hinder the progress of individuals to become highly literate and think critically? I don't have the answers to these questions, but from the point of view of a teacher, my purpose was to raise the question. My central concern is how the development of cognitive functions and consciousness in the phases of language acquisition and literacy will be effected by the use of LLM AI in classrooms. Will we remain motivated to acquire language, reading and writing skills, in the way we do today? (some would argue ,we do not already pay enough attention to developing these skills, or do it will enough); or will we find ways of using LLM AI to enable individuals (especially in the crucial phases of childhood), to become more literate, and better, critical thinkers? (I think there will doubtless be uses, but they need to be refined and strategic). My experience with undergraduate students suggests that even at the later stages of literacy development, there is still a lot that can be done to develop higher levels of literacy and critical thinking. Indeed, I feel literacy is a lifelong journey none of us complete. Unfortunately, not all - and I think most - of the ways in which LLM AI is currently being used by students, isn't contributing towards the development of their literacy and critical thinking. There are, for sure, positive uses, and I would love to have had more time to explore both negative and positive uses in greater detail - but with this talk, I ventured to suggest concerns I was hearing among my community of educators, but hadn't seen fully articulated at the time. Thanks again for your brilliant and thoughtful response.

I worry that the inner searching for optimal words, the between the ears composition of phrases, sentences, paragraphs and pages, will be overlooked and eventually lost to the easy rote transcription of ideas by a robotic writer. Along with composition of sentences comes THOUGHT and critical analysis of issues; both of which are “mammalian,” “human,” and vital to individuality. I wonder and worry about AI writing like Kerouac, Shakespeare or Tolstoy. How would AI interpret the writings of Kabir , versus how Robert Bly transcribed the thoughts?

lewis mcnulty

SAME

GPT 5: Hold my beer.

naitik what the bludclart

naitik i love you so much it gets me errect🎉🎉🎉

I wrote the talk. I'm so encouraged to see that you recognize how learning is about the process and not the product. We have a great future as long as people like you continue to think that way! ☺

was it better before him?

Hi Larushka. Thanks for watching! I wrote the talk. I'm not sure what you mean by the comment, but I can assure you all the facts in my talk are accurate. You may be referring to the talk sub-title - 'Why Chat-GPT Can't Write For You" - which is not a sub-title I gave the talk. My central claim is that we don't know how AI will effect the cognitive benefits of language learning and literacy development. The ambition of the talk is to start a conversation about that - which lots of people here are engaging in - which is great! Of course, ChatGPT and many software language tools can help with the technical clarity of our writing, but on a deeper level, the use of language and writing are integral to the act of thinking - so changes to the way we think need really carefully scrutiny and consideration.

The header was definitely misleading. I’m an IT tech and educator working specifically with seniors and this is a great tool for them - used safely and responsibly. Thanks for your talk!

@@larushka1 I can imagine how useful these tools are for seniors who already have basic levels of literacy or face challenges maintaining their cognitive faculties. Great work!

I went ahead and here is the AI summarization: This TED talk is given by David Savill, a teacher of reading and writing and a novelist. He argues that AI will never be able to replace the importance of human literacy and critical thinking. The speaker begins with a story about Paulo Freire, a Brazilian teacher who was jailed for teaching literacy to agricultural workers. Freire believed that literacy is a powerful tool that can empower people and give them a voice. Savill argues that literacy is not just about being able to read and write, but also about being able to think critically and creatively. Savill then discusses the recent development of large language models, such as ChatGPT, which can generate human-quality text. He worries that these models will lead people to believe that they don't need to learn how to read and write for themselves. Savill argues that using AI to write is not the same as writing yourself. When you write, you are using your brain to process information and generate ideas. This process helps you to develop your own unique voice and perspective. Savill concludes his talk by arguing that literacy is essential for human flourishing. It allows us to explore our own consciousness and to make sense of the world around us. As we enter the age of artificial intelligence, it is more important than ever for us to cultivate our own humanity. Here are the key points: Literacy is a powerful tool that can empower people and give them a voice. AI can generate human-quality text, but it cannot replace human literacy. Writing is a process that helps us to develop our own unique voice and perspective. Literacy is essential for human flourishing.

First reply to the first comment

The saddest part is that you not only admit wilful ignorance, but that if you had watched it to its conclusion you would have heard what the questions were and why they can’t be answered by a machine. It’s comments like yours that make me understand how the world you describe would come about, not by those who seek to do harm, not by those who fight it, but by those who are apathetic and lazy.

@@jenniferelyse5134 my best friend and, probably, the smartest person I’ll ever know was massively dyslexic. He would speak to me what he was going to write and it was sheer brilliance. He would lock himself away and write for days and produce unreadable nonsense. Were he still alive, voice to text and AI would allow him to share his towering genius with the world I, on the other hand, love writing and would never want words that weren’t mine to be my sword and shield. Making art with computers, however, is simply an exercise in frustration. For me, AI image generation is an absolute boon For those who wish to scour their soul: write, draw, paint, or dance. For those who wish a product that is kind to their precious time on this planet: AI ChatGPT can write for many people. Writing isn’t the only way to know yourself

@@ScorpioIsland Hello both. I wrote the talk. Thankyou for watching. This is an interesting conversation because you both make points I made in the talk itself. In the middle section, I talk about my profoundly dyslexic daughter, and the dangers of being a literacy snob - so I agree with you entirely. The subtitle 'Why ChatGPT can't write for you' was added by TED, and is not a claim I make in the talk, although it is apposite if we consider writing as analogous with a certain kind of thinking. My concern is about literacy in education - which is my field. The process of learning to read and write develops cognitive skills beyond the product of reading and writing. Each generation learns in different ways, and their cognitive skills are shaped in different ways. We simply don't know how those cognitive skills will be effected if we allow LLM AI to become a part of the literacy process. There could be benefits, but there could also be dangers. The second point is historical - literacy as we know it was one of the major factors in the development of advanced economies, and democracy in particular. This is, in part, because one of the cognitive effects of literacy is advanced critical thinking. The purpose of my talk is to raise questions and create a conversation about how changes in the process of language acquisition, reading and writing, will effect the connection between self and society, and the critical thinking ability of humans. No-one knows the answers to these questions, but the first step is to ask them.

Interesting point of view! In the talk, I praise the positive uses of AI and embrace change. I think it isn't contradictory to also suggest we closely guard the cognitive benefits of learning to read and write well.

@@persimian4997 there's many other cognitive benefits like learning to reason or do math. one can wish to be accomplished in everything, even playing the accordion but there are trade offs from time and energy. and surely, i hope he's not dictating what is best for everyone, like 1984. what he stated is more of a reflection of his bias without consideration for the opportunity cost. what he hoped to say was profound was not.

I don't disagree. There are some great creative uses of AI. I'm already attending conferences exploring them. The concern I express in my talk is about what we might lose if we cede too much control to AI in the specific field of literacy education, and how large language models challenge what we mean by the literate citizen. There are definite opportunities in large language model generative AI systems, but drawbacks to be aware of too. Personally speaking, I can say I'm not worried about losing my job! 😂

@@persimian4997 Hello thank you for the very thoughtful response. LLMs are giving us an amazing microscope into language (human and non-human), and they force us to examine questions like what literacy is and what use it actually has. Literacy as it's defined now might not be as important as it has been in the past. These new tools may give us the *choice* of literacy (or not), which IMO is an incredible superpower. Powerful tools require *respect*, not fear. Instead of focusing on the few possible harms, why not focus on the myriad of ways new forms of literacy can be helpful? As for jobs -- if literacy as we know it is going away, then any literacy education jobs (like yours) will also go away. No worries -- there will probably be something else to do. :) Cheers

This is pretty much like I write technical articles