A lot of people are asking me for my opinion of ChatGPT these days. Here is what I am telling them:
(1) What I think – I enjoy using it
I am really enjoying using it – even though it is imperfect – it is a useful cognitive tool that makes a lot of mistakes, but it will get better over time. I have to spend a lot of time checking assertions to determine if they are actually true or not (see addendum below). That is one of its biggest weaknesses. However, it will get better over time.
(2) Can it be viewed as a virus? See reference [1]
Yes. A virus spreads, mutates, and can cause harm. So yes, in some ways it is like a virus. However, people can adapt to viruses. Think of it more like a bacteria. Some are harmful, and some are helpful – and in fact we would not live long without all the helpful bacteria in our body. To deal with complex urgent problems, we need better tools – and ChatGPT-like systems will certainly get better over time (less misinformation generated from inferior models), and people will adapt to the negative parts (harms), while embracing the positive parts (benefits).
(3) Should OpenAI have released ChatGPT?
Yes, AI will continue getting better, and the world needs to ‘upskill with AI’ (see reference [2]) – and get prepared. Future generations will have 100 digital workers working for them, and they need to be able to imagine a worthy future to be collectively working towards with AI/cognitive tools, assistants, collaborators, coaches, and eventually cognitive mediators (trusted to negotiate contracts on your behalf) (see reference [3]). Will bad things happen along the way as well?. Yes, that is true of all technologies – the potential for harm and good are always both there. I want to live in a world where we have harnessed AI primarily for its benefits, while mitigating the potential harms. Yes, regulations have arrived, and more will come in time as more cases of harm appear. Unfortunately, we are not smart enough to know the future as Kedrosky mentioned (see reference [1]).
(4) What is ChatGPT? A ‘humanity-level intelligence test’
I use ChatGPT to work on problems that are important to me. More broadly, I see ChatGPT as a ‘humanity-level intelligence test.’ Millions of people are already fantastic at generating misinformation using incorrect or outdated models of the world. We need better models. That is the theme in my recently co-authored book with Maglio, Vargo, and Warg – ‘Service in the AI Era’ (see reference [4]) – that we need betters models of the world (science), better models in people’s heads (logics), better models of organizations (architectures, including businesses, universities, cities, nations), and last but not least better AI models. We also need humanity-centered design to build a better world (see reference [5]). Yes, I see this as continuing to evolve all disciplines that are on a transdiscipliny journey, including the emerging transdiscipline of service science (see reference [6]).
(5) What do I recommend for the future? Learn to invest wisely and systematically
Try ChatGPT, and think of it today as a cognitive muse (see reference [7]), that will eventually become a useful cognitive tool, assistant, collaborator, coach, and finally mediator (see reference [3]). Overall, humanity needs to learn to invest in improved win-win interaction and change processes (service innovations) (see pages 45-54 in “Service in the AI Era” – reference [4]). On the current trajectory with increasing AI capabilities, we can get there by having “AI/digital twins” of all responsible actors (people, businesses, universities, cities, nations, etc.) that both give and get service daily, and try to invest in a positive future. This will allow calculating benefits and harms better than we can do today. As individuals adopt better strategies based on better models, collectively the future is shaped with better models. All models are wrong, but some are useful. Better models are more useful for shaping a positive future. This will allow shaping the future (not predicting it – that is impossible) better than we can today – shaping the future with more benefits and fewer harms – more good days of service for more people, and fewer bad days of service for just a few people.
(6) What do I recommend for today? Spend one hour ‘upskilling with AI’
To spend one hour ‘upskilling with AI,’ I recommend watching some social media and readings – see these YouTubes and Tweets (see references [8], [9], [10], [11], and [12]). Remember ChatGPT-types systems are a mirror of the information people share with each other daily. With better models (science, logic, architecture, AI) the quality of information sharing can potentially increase substantially in the future.
Hope this perspective is helpful to some readers. -Jim
Addendum
Spohrer: “Please describe how AI systems, such as ChatGPT that generate text responses to prompts, can help users evaluate fact from fiction, and well-sourced opinions from non-expert influencer opinions.”
ChatGPT: “AI systems such as ChatGPT are trained to generate text responses based on a given prompt. In the context of helping users evaluate fact from fiction and well-sourced opinions from non-expert influencer opinions, ChatGPT could be used to generate responses that highlight the credibility of a particular source or piece of information. For example, if a user were to ask ChatGPT whether a certain statement was true or false, ChatGPT could generate a response that indicates whether the statement is supported by evidence or whether it is merely an opinion. Additionally, ChatGPT could be used to generate responses that provide information about the credentials and expertise of a particular person or organization, which could help users determine the reliability of their opinions.”
Spohrer: “Provide additional feedback: This response is very helpful. In fact it addresses the key weakness that I see in ChatGPT. Furthermore, this type of response annotation would make a useful sidebar for many responses. See more thoughts here: https://service-science.info/archives/6309”
For those interested in exploring some of the dangers inherent in powerful AI technologies, see reference [13]. In [13], Marcus summarizes three key problems with these types of systems, and proposes four remedies that should be explore. The fourth remedy seems very important to me – build better AI to verify truth. For more examples of misinformation (old information, made-up information, etc.) see reference [14]. Nevertheless, with the help of [13] and [14] to keep our eyes wide open about potential harms, Foundational AI models and systems such as OpenAI’s ChatGPT, which achieved 1 million users in the first five days of its release to the general public, seem to be at or near a tipping point of being generally useful to people across a wide range of industries, enabling ‘the possibility of human-machine hybrid work’ [15].
References
[1] BiblioL2022 Loizos C (2022) Is ChatGPT a ‘virus that has been released into the wild’? Government & Policy. TechCrunch. URL: https://techcrunch.com/2022/12/09/is-chatgpt-a-virus-that-has-been-released-into-the-wild/ Quotes: “Paul Kedrosky isn’t an educator per se. He’s an economist, venture capitalist and MIT fellow who calls himself a “frustrated normal with a penchant for thinking about risks and unintended consequences in complex systems.” But he is among those who are suddenly worried about our collective future, tweeting yesterday: “[S]hame on OpenAI for launching this pocket nuclear bomb without restrictions into an unprepared society.” Wrote Kedrosky, “I obviously feel ChatGPT (and its ilk) should be withdrawn immediately. And, if ever re-introduced, only with tight restrictions.” We talked with him yesterday about some of his concerns, and why he thinks OpenAI is driving what he believes is the “most disruptive change the U.S. economy has seen in 100 years,” and not in a good way.”; “I heard from a colleague at UCLA who told me they have no idea what to do with essays at the end of the current term, where they’re getting hundreds per course and thousands per department, because they have no idea anymore what’s fake and what’s not. So to do this so casually — as someone said to me earlier today — is reminiscent of the so-called [ethical] white hat hacker who finds a bug in a widely used product, then informs the developer before the broader public knows so the developer can patch their product and we don’t have mass devastation and power grids going down. This is the opposite, where a virus has been released into the wild with no concern for the consequences.”;
[2] BiblioS2022 Spohrer J (2022) The Future of AI and IA: A Service Science Perspective. Presentation to India’s AIMA Delegation at Berkeley on Monday November 28, 2022. URL: https://www.slideshare.net/spohrer/spohrer-aima-20221128-v8pptx Quotes: ” Still haven’t found a good way to include these materials from Frank, Thompson, Davis, and Zyer in my standard stump speech about the future of AI and IA (Intelligence Augmentation – AKA “Learning to Invest in Upskilling with AI”) which is the main message of my recent book “Service in the AI Era” (pages 45-54).”; “The Future of AI and IA: A Service Science Perspective: This talk explores the future of Artificial Intelligence (AI) and Intelligence Augmentation (IA). IA can initially be thought of as people “upskilling with AI” tools. In this talk, after a preamble on understanding “value” in the world today, a roadmap for solving AI is proposed. Second, an approach to measuring IA is suggested. Finally, the importance of trust, purpose, and human values are examined as “reality” is extended for populations of evolving service system entities learning to invest systematically in becoming better future versions of themselves. For example, AI, both service robot automation and service augmentation platforms, are poised to improve service productivity, quality, compliance, sustainable innovation, resilience, equity and inclusion for under-served populations. Service is defined as the application of knowledge for the benefit of another. Service innovations improve interaction and change processes in business and society. However, to achieve these outcomes and create value with AI, responsible actors (people, businesses, governments, universities – AKA service system entities) must learn to invest wisely in becoming better future versions of themselves augmented by their AI digital twin. Learning to invest systematically in upskilling with AI can accelerate both value cocreation and capability coelevation in a virtual cycle of responsible actors’ improved win-win interaction and change processes. Service innovations improve win-win interaction and change processes in business and society, but require learning to invest systematically and wisely in shifting conflict and fear to understanding and more than coexistence to collaboration. However, great risks must also be avoided – as benefits for some may cause harm to others. ”
[3] BiblioS2017 Spohrer J, Siddike MAK (2017) The future of cognitive systems: tool, assistant, collaborator, coach, mediator. In Augmented Intelligence: Smart Systems and the Future of Work and Learning (New Literacies and Digital Epistemologies), Editor D. Araya. Peter Lang Publisher. URL: https://www.amazon.com/Augmented-Intelligence-Learning-Literacies-Epistemologies/dp/1433133334 Quotes: “Chapter Abstract – Abstract: In this chapter we explore the future of cognitive systems that progress from tool to assistant to collaborator to coach to mediator over time, augmenting both cognitive and social intelligence. We argue that in the decades to come most people will use cognitive mediators to boost win-win interactions with others. Furthermore, we argue that future innovation will be driven by people with a T-shaped skillset and mindset, who with the help of their cognitive mediators, know how to rapidly rebuild socio-technical systems from scratch. T-shaped professionals with cognitive mediators will be expert at people-centered system redesign thinking, and have a deep understanding of service science – the science of value co-creation and capability co-elevation.”; “Book – Where the Agricultural Revolution harnessed domesticated animals for pastoral farming, and the Industrial Revolution leveraged machines for factory production, so today the Computational Revolution is advancing computers to augment human intelligence. Indeed, many now argue that the promise of exascale computing and the slow migration towards a computational society may represent a new threshold in human history. This “transcension” of earlier stages of tool-mediated work and learning foreshadows a momentous change in the kinds of cities we might build, the kinds of medicine we might practice, and the kinds of education we might provide. What is perhaps most surprising about the current Computational Revolution, however, is its expanding reach. The question that many now ask is “what is the trajectory of this human-machine symbiosis?” It would appear that we are on the cusp of a sea change in our capacities to augment human intelligence. But what is the future of work and learning? Will augmented intelligence help us in transforming a waning industrial society? These are the kinds of questions that we explore in Augmented Intelligence: Smart Systems and the Future of Work and Learning.”
[4] BiblioS2022 BiblioS2022 Spohrer J, Maglio PP, Vargo SL, Warg M (2022) Service in the AI Era: Science, Logic, and Architecture Perspectives. Business Expert Press. URL: https://www.amazon.com/Service-AI-Era-Architecture-Perspectives/dp/1637423039/ Quotes: “Are you prepared for the coming AI era? AI advances will profoundly change your daily service interactions, so this book provides readers with a necessary understanding of service, the application of resources (e.g., knowledge) for the benefit of another. In just minutes, you can learn about today’s use of early-stage AI for automation and augmentation, and essential elements of service science, service-dominant (S-D) logic, and Service Dominant Architecture (SDA). Ultimately improved service for all is possible with human-level AI and digital twins – but requires investing wisely in better models: Better models of the world both complex natural and social systems (science), better mental-models in people to improve interactions (logic), better cultural and structural models of organizations to improve change (architecture), and better trusted and responsible AI models. The service innovation community studies and builds better models to improve interactions and change in business and society. The book challenges all responsible actors – individuals, businesses, universities, and governments – to invest systematically and wisely to upskill with AI (the X+AI vision). The service innovation community is a growing transdiscipline harnessing all disciplines to become better T-shaped professionals. Extensive end notes, bibliography, and index are provided.”;
[5] BiblioN2023 Norman DA (2023) Design for a Better World: Meaningful, Sustainable, Humanity Centered
URL: https://www.amazon.com/Design-Better-World-Meaningful-Sustainable/dp/0262047950 Quotes: “Human behavior brought our world to the brink, human behavior can save us. The world is a mess. Our dire predicament, from collapsing social structures to the climate crisis, has been millennia in the making and can be traced back to the erroneous belief that the earth’s resources are infinite. The key to change, says Don Norman, is human behavior, covered in the book’s three major themes: meaning, sustainability, and humanity-centeredness. Emphasize quality of life, not monetary rewards; restructure how we live to better protect the environment; and focus on all of humanity. The book presents an eye-opening diagnosis of where we’ve gone wrong and a clear prescription for making things better. Norman proposes a new way of thinking, one that recognizes our place in a complex global system where even simple behaviors affect the entire world. He identifies the economic metrics that contribute to the harmful effects of commerce and manufacturing and proposes a recalibration of what we consider important in life. His experience as both a scientist and business executive gives him the perspective to show how to make these changes while maintaining a thriving economy. Let the change begin with this book before it’s too late”;
[6] BiblioW2022 WladawskyBerger I (2022) The Current State of Service Science. URL: https://blog.irvingwb.com/blog/2022/12/reflections-on-the-current-state-of-service-science.html Quotes: “On October 17, the National Academy of Engineering (NAE) conducted an online forum on Service Systems Engineering in the Era of Human-Centered AI. “With AI advances poised to drive service system productivity and quality – similar to the way previous generations of technology revolutionized agriculture and manufacturing productivity and quality – it is time to take stock for industry-academic-and-government stakeholders on this important topic,” wrote the NAE in its website.”; “Sometime in 2016 I had an interesting conversation with analysts from an IT research organization who were preparing a report on the state of service science. They noted that we were hearing quite a bit less about service science in those days compared to 5 or 10 years earlier. Was it because we had become tired of the subject and moved on to other areas of innovation?”; “For example, in their recent book Service in the AI Era, Jim Spohrer, Paul Maglio, Stephen Vargo, and Marcus Warg wrote that “Service is quickly becoming the central concept of our time, as service offerings become infused with advanced technologies like artificial intelligence (AI) and scale to new levels of quality, productivity, compliance, and sustainable innovation. Throughout history, scientific revolutions have been launched when new tools make possible new measurements and observations, e.g., the telescope, the microscope, spectrometers, DNA sequencers. Our new big data tools have now been ushering an information-based scientific revolution, helping us extract insights from the huge amounts of data we’ve been collecting by applying tried-and-true scientific methods, that is, empirical and measurable evidence subject to testable explanations and predictions.”
[7] BiblioF2022 Flux Collective (2022) A muse, rather than an assistant. The FLUX Review, Ep. 64 August 18th, 2022. FLUX Review. URL: https://read.fluxcollective.org/p/64 Quotes: “A muse, rather than an assistant. With large-language-model-powered tools taking early adopters’ attention by storm, we are inclined to pause and consider the potential future uses of these tools. Some of them already exhibit remarkable abilities. Still, rough edges show up quickly: faces melt, sentences stop making sense, and so on. We are still in the early stages of understanding how to best use these tools. As we continue to explore, it is important to keep in mind both the potential benefits and their potential risks. However, we have found one use that already seems intriguing: a source of inspiration. Especially in the busy Midjourney community, the creative vibe is nearly palpable. People riff on each other’s creations, showcase their best, and most interestingly, show fascinating new uses of the tool’s output.”;
[8] BiblioF2022 Frank T (2022) Notion AI is Here: 10 Ways to Use This Powerful New Tool URL: https://youtu.be/0DIn0Ws9yTE Quotes: “Notion just released a suite of AI tools that live directly inside the Notion app. Today I’ll show you 10 amazing examples of what these artificial intelligence tools can do – from generating blog posts and comparison tables to even writing code! Learn more about Notion AI and join the waitlist: https://www.notion.so/product/ai”; “Written version of the video: https://thomasjfrank.com/notion-ai-guide/”
[9] BiblioT2022 Thompson AD (2022) Devoxx Keynote – Artificial Intelligence: You Are Here – Alan D Thompson (12/Oct/2022. URL: https://youtu.be/D3tTsoX02d8 Quotes: “On 20221015 – 6,303 views Oct 13, 2022 The Memo: https://lifearchitect.ai/memo/ Read more: https://lifearchitect.ai/ https://lifearchitect.ai/models/ A look at Transformer, RoBERTa, GPT-3, Pathways + PaLM, Chinchilla, Google Imagen, Google Parti, NUWA-Infinity, Google Imagen Video, and much more… Dr Alan D. Thompson is a world expert in artificial intelligence (AI), specialising in the augmentation of human intelligence, and advancing the evolution of ‘integrated AI’. Alan’s applied AI research and visualisations are featured across major international media, including citations in the University of Oxford’s debate on AI Ethics in December 2021. https://lifearchitect.ai/”;
[10] BibloD2022 Davis T (2022) TL;DR 11 free AI tools that will make you unstoppable. URL: https://twitter.com/tessardavis https://twitter.com/TessaRDavis/status/1595749323995840513 Quotes: “TL;DR 11 free AI tools that will make you unstoppable.• #1: Super Meme
• #2: Hotpot
• #3: Dall•E 2
• #4: Sembly
• #5: Browse AI
• #6: Posied
• #7: Lalal
• #8: Magic Eraser
• #9: Synthesia
• #10: Soundraw
• #11: Replika”;
“#1: Make custom memes with Super Meme.
Lacking meme inspiration?
http://supermeme.ai
• Write a sentence
• Describe an emotion
• Upload your own image if you want
Your own custom meme in 2 minutes.”;
“#2: Restore old pictures with Hotpot.
Have an old photo you’d like to improve?
Upload it to http://hotpot.ai
It’ll:
• Remove scratches
• Sharpen images
• Repair gaps
Transforms your damaged photos.”;
“#3: Create amazing images with Dall•E 2
Have an idea in your head for an image?
http://openai.com
• Type in the description
• Be as specific as you can
• It’ll search for existing images
• Or generate new ones for you now
The AI stock photographer you always wanted”;
“#4: Easily take notes with Sembly.
Need meeting notes?
http://Sembly.ai
• Generates meeting summaries
• Automates your minutes
• Shares with your team
Integrates across all platforms.”;
“#5: Check out your competition with Browse Ai.
Need info on your rivals?
http://Browse.ai
• Scrape posts from Medium
• Get lists of YouTube videos
• Extract company info from LinkedIn
Make your research easy.”;
“#6: Improve your public speaking skills with Poised.
Are you a nervous presenter?
Get coaching with http://posied.com
• Record
• Get feedback
• Track your progress
Watch your skills improve.”;
“#7: Separate audio tracks with Lalal.
Have a music track you want to separate?
http://lalal.ai
Just upload the track
It extracts the vocals and separates each instrument.
No need for fancy editing tools.”;
“#8: Edit out parts of photos with Magic Eraser.
Have a great photo but with something annoying in the background?
Remove it easily:
http://Magiceraser.io
• Upload the photo
• Select the part you don’t want
• Press ‘erase’
And voila. It’s gone.”;
“#9: Create professional videos with Synthesia.
Need a video quickly?
http://synthesia.io
• Add your text
• Use templates
• Edit it as needed
You’ll get an AI-generated video within minutes.”;
“#10: Create unique music with Soundraw.
Need some music for your video?
http://soundraw.io
• Choose your mood
• Choose the genre
• Choose a length
• Edit if needed
Now you’ve got your own AI music.”;
“#11: Create a new friend with Replika.
Feeling lonely?
http://replika.com
• Build your own AI
• The more you chat the more they learn
• Express yourself and build a human-AI bond
Is this the future?”;
[11] BiblioZ2022 Zyer E (2022) If you’re not using AI, you’re losing out.8 mindblowing AI tools (you can use today) that will save you hours. Twitter Eyisha @eyishazyer. URL: https://twitter.com/eyishazyer/status/1597573909104427009 https://twitter.com/eyishazyer/status/1597573909104427009?s=20&t=v3okqE9sQNZTEmJocroSTQ Quotes: “”;”1. @lexdotpage
Unlock your best writing with Lex.
Any time you don’t know how to continue your text, type +++, and Lex continues for you.
It also helps generate titles for your texts!
“;
“2. @runwayml
The runway is the content creation suite of the future. It has magical AI tools like:
– change images with text descriptions
– remove objects in videos
– remove video backgrounds
– expand images with descriptions (as seen in the video)
+++”;
“3. @diffusionbee
DiffusionBee is a free macOS app for Stable Diffusion. Give it a text prompt, and it generates a picture based on your text.”;
“4. @LexicaArt
Providing prompts for text-to-image solutions is an art.
LexicaArt is a Stable Diffusion search engine that shows prompts others have used to generate images.
You can also see variations of a prompt.”;
“5. @metaphorsystems
Metaphor is a new search
engine based on generative AI.
You prompt it by writing a phrase that looks like it could end with a link.”;
“6. @SoundrawUS
Do you need some unique music for your podcast or video? Soundraw is a music generator for creators.
Select the type of music you want – genre, instruments, mood, length, etc. – and let their AI generate beautiful songs and variants for you.”;
“7. @clipdropapp
Relight your existing photos
& drawings with ClipDrop Relight.
Relighting a picture after it is captured is possible thanks to their custom AI that maps depth into your pictures.”;
“8. Talk to books
Get quotes from more than 100 000 books that respond to your question.
A creativity tool by Google to explore new ideas and get relevant quotes.
http://books.google.com/talktobooks/”;
[12] BiblioG2022 Griffith T (2022) AI-Supported Tools for Personal Automation. Website. URL: https://terrigriffith.com/blog/ai-supported-tools-for-personal-automation Quotes: “AI-Supported Tools for Personal Automation – Automation is part of our work. The question is whether we are finding the tools to augment our abilities or waiting for our organizations to decide what to automate and who to replace. Research finds that people who craft their work are more productive and employable. In this earlier post with Brett Li of Tonkean, we talk about that research and the steps you can take to craft your work. But what tools can help us?”;
[13] BiblioM2022 Marcus G (2022) AI’s Jurassic Park moment. New systems like chatGPT are enormously entertaining, and even mind-boggling, but also unreliable, and potentially dangerous. The Road to AI We Can Trust. Substack – Gary Marcus Dec 10, 2022. URL: https://garymarcus.substack.com/p/ais-jurassic-park-moment Quotes: “Something incredible is happening in AI right now, and it’s not entirely to the good. Everybody is talking about systems like chatGPT (OpenAI), Dall-E 2, and Lensa that generate text and images that look remarkably human-like, with astonishingly little effort. These systems can be incredibly fun to play with. Take this example, generated by chatGPT by Henry Minsky (son of Marvin Minsky, one of AI’s founders), who asked chatGPT to “Describe losing your sock in the dryer in the style of the declaration of independence”:”; “That a machine could do this, with so little effort on the part of the user, is frankly mind-boggling. But at the same time it is, or should be, terrifying. It is no exaggeration to say that systems like these pose a real and imminent threat to the fabric of society. The core of that threat comes from the combination of three facts:”; “Spohrer: (1) Inherently unreliable – user beware, (2) Can be weaponized, (3) Inexpensive to bad actors.”; “Thanks Gary for making helpful points that these Generative-Pretrained-Transformer AI systems, like ChatGPT, are simultaneously very fun to use and yet (1) make many mistakes, so user beware, (2) can be weaponized by bad actors, and (3) are inexpensive to use by bad actors and other users alike. My further opinion here https://service-science.info/archives/6309”; “YouTube: Dr. Ian Malcolm gives Hammond an ethics lecture: Gee the lack of humility before nature that is being displayed here staggers me. Don’t you see the danger inherent in what you are doing here. Genetic power is the most awesome force the planet has ever seen. But you wield it like a kid who has found his dad’s gun. Scientific power did not require any discipline to attain. Didn’t earn it. Didn’t take responsibility for I’ll tell you the problem with the scientific power that you’re using here: it didn’t require any discipline to attain it. You read what others had done and you took the next step. You didn’t earn the knowledge for yourselves, so you don’t take any responsibility for it. You stood on the shoulders of geniuses to accomplish something as fast as you could and before you even knew what you had you patented it and packaged it and slapped it on a plastic lunchbox, and now you’re selling it, you want to sell it!”; “Marcus: All of this raises a critical question: what can society do about this new threat? Where the technology itself can no longer be stopped, I see four paths, none easy, not exclusive, all urgent:”;”Spohrer paraphrase: (1) social media and search engine bans (like StackOverflow), (2) nations need to make misinformation an actionable offense, like libel, (3) ensure provenance of all user accounts (real people, real responsibility), (4) build better AI to verify truth.”;
[14] BiblioB2022 Banavar G (2022) ChatGPT’s deep fake text generation is a threat to evidence-based discourse. Medium – Guru Banavar. URL: https://medium.com/@guruduth.banavar/chatgpts-deep-fake-text-generation-is-a-threat-to-evidence-based-discourse-c096164207e0Quotes: “The insidious nature of such generative AI models is that they produce plausible sounding health information in a smooth conversational style, that is at best is generic in nature and at worst completely fictional. The problem is that most people cannot tell the difference between the two. “; “It can be argued that the information put out by ChatGPT is generic conventional wisdom, so it is not seriously problematic, and that it is the nature of science to continuously unearth new and improved truths. Also, it will always be the case that there will be outdated information in vogue. After all, the medical practices from the last few centuries are still being practiced in some form by subgroups of people somewhere in the world. But this is precisely why we would like modern AI systems to help us in curating the most evidence-based and trust-worthy information!”;
[15] BiblioM2022 Mollick E (2022) ChatGPT Is a Tipping Point for AI. HBR December 14, 2022. Via: Larry_Hiner URL: https://hbr.org/2022/12/chatgpt-is-a-tipping-point-for-ai Quotes: “Summary. We’re hitting a tipping point for artificial intelligence: With ChatGPT and other AI models that can communicate in plain English, write and revise text, and write code, the technology is suddenly becoming more useful to a broader population of people. This has huge implications. The ability to produce text and code on command means people are capable of producing more work, faster than ever before. Its ability to do different kinds of writing means it’s useful for many different kinds of businesses. Its capacity to respond to notes and revise its own work means there’s significant potential for hybrid human/AI work. Finally, we don’t yet know the limits of these models. All of this could mean sweeping changes for how — and what — work is done in the near future.”; “This highlights the third major change that happened with this release: the possibility of human-machine hybrid work. Instead of prompting an AI and hoping for a good result, humans can now guide AIs and correct mistakes. (Despite what my AI writing partner claims above, it’s not always accurate.) This means experts will be able to fill in the gaps of the AI’s capability, even as the AI becomes more helpful to the expert. This sort of interaction has led to increases in performance of players of Go, one of the world’s oldest and most complex games, who have learned from the AIs that mastered the sport, and become unprecedentedly better players themselves.”; “The problems of AI remain very real, however. For one, it is a consummate bullshitter, and I mean that in a technical sense. Bullshit is convincing-sounding nonsense, devoid of truth, and AI is very good at creating it. You can ask it to describe how we know dinosaurs had a civilization, and it will happily make up a whole set of facts explaining, quite convincingly, exactly that. It is no replacement for Google. It literally does not know what it doesn’t know, because it is, in fact, not an entity at all, but rather a complex algorithm generating meaningful sentences.”;
