🚀 The Economist tested both Google's Bard & ChatGPT: see the results ; how to use ChatGPT to boost your writing, & more
ChatGPT passes exams from law and business schools, helps write and direct a film
Bonjour,
Vous recevez la newsletter Parlons Futur : une fois par semaine au plus, une sélection de news résumées en bullet points sur des sujets tech 🤖, science 🔬, éco 💰, géopolitique 🌏 et défense ⚔️ pour mieux appréhender le futur 🔮.
Je m'appelle Thomas, plus d'infos sur moi en bas d'email.
Voici donc ma dernière sélection très axée ChatGPT cette semaine, mais pas que ! Enjoy !
L’apéro
OpenAI's ChatGPT is now the fastest-growing app in human history, reaching an estimated 100 million active monthly users in just the 2 months since its November release. (source)
Cette semaine Google (Alphabet) présentait BARD, son ChatGPT maison, lors d’une conférence de presse. Pendant la démonstration, le robot donne une mauvaise réponse à une question sur le premier observatoire à avoir pris des images d’une planète hors du système solaire. L’action a perdu dans la foulée 8% soit 100Mds$. (TTSO et Numérama)
à noter aussi : Google investit 300 millions de dollars dans Anthropic, développeur d'un concurrent de ChatGPT, fondée par des anciens d'OpenAI (source)
Google has a model that can generate high-quality multi-minute music from text descriptions: "a calming violin melody backed by a distorted guitar riff". Unlike OpenAI and CHatGPT, Google isn't making it public just yet (listen to examples here)
Si vous l'avez ratée, la dernière vidéo d'une minute de robot de Boston Dynamics, toujours aussi impressionnant !
Tech analyst Ben Evans: "I really wish Meta would stop promoting VR with images of people using half a dozen giant floating virtual screens. The future of work is not about seeing more columns in your spreadsheet. It’s about smaller screens, less data on display and more intelligence and workflow
Ben Evans on other unrelated topics:
What a cashless society looks like: there wasn't a single bank robbery last year in Denmark, a country of close to 6 million people (source)
IBM has filed more patents than anyone else in the USA for 29 years running. This is one reason no-one in tech thinks patents tell you anything - IBM hasn’t done anything interesting for at least that long. But now, it says, it will shift away from ‘numerical patent leadership’.(source)
A startup developed a bot leveraging ChatGPT able to negotiate a bill: "Our DoNotPay ChatGPT bot talks to Comcast Chat to save one of our engineers $120 a year on their Internet bill." (voir la négo ici)
Et quand les entreprises feront pareil en face, ça va devenir comique...
La start-up française PASQAL est un des leaders mondiaux de l’informatique quantique
Co-fondée par Alain Aspect, Prix Nobel de physique 2022, elle vient de lever 100M€ pour accélérer son développement.
Dès aujourd’hui, elle permet à ses clients BASF, Crédit Agricole, BMW, et d’autres sociétés du Global Fortune 500, de dépasser les limites de l’informatique classique et d’avoir accès à des puissances de calcul jamais envisagées auparavant. (TTSO et BFMBusiness)
New AI tool can transform existing videos into new ones by applying any style specified by a text prompt or reference image. (watch the 2-min demo, quite amazing)
Wharton professor Ethan Mollick: "Here is an AI-generated person with a (now realistic) AI-generated voice delivering the beginning of a lecture where all of the content & slides are made by AI." Watch the 1-min video
Parlons passé : Recent research shows that the Tyrannosaurus rex had as many neurons in their brains as monkeys do (source)
With that many neurons, a T-Rex wouldn't have just possessed unexpected cognition. It also might have lived longer, up to 40 years.
That's enough time and smarts to potentially be a social creature with its own culture, like primates and whales, and also suggests they may have worked together, too.
440 years of progress in shipping:
English merchant fleet, year 1582: capacity 68,000 tons, crew of 16,000 sailors
One container ship today: 236,228 tons, crew of 22 (source)
This AI tool lets you virtually try 40 hairstyles for 15€
À votre bon coeur ❤️
Si vous appréciez cette synthèse gratuite, n’hésitez pas à prendre 3 secondes svp pour l’envoyer à ne serait-ce qu’un contact 🙂
Et si on vous a partagé cet email, vous pouvez cliquer ici pour vous inscrire et ne pas manquer les prochains
À table !
OpenAI CEO Sam Altman’s latest remarks on ChatGPT (Forbes)
One of the things is that the base model for ChatGPT had been available for a long time, you know, like 10 months, or whatever. [Editor’s note: ChatGPT is an updated version of model GPT-3, first released as an API in 2020.] And I think one of the surprising things is, if you do a little bit of fine tuning to get [the model] to be helpful in a particular way, and figure out the right interaction paradigm, then you can get this success. It's not actually fundamentally new technology that made this have a moment. It was these other things. And I think that is not well understood. Like, a lot of people still just don't believe us, and they assume this must be GPT-4.
Is OpenAI still open enough? :
I think the most important way we do that is by putting out open tools like ChatGPT. Google does not put these things out for public use. Other research labs don't do it for other reasons; there are some people who fear it’s unsafe.
But I really believe we need society to get a feel for this, to wrestle with it, to see the benefits, to understand the downsides. So I think the most important thing we do is to put these things out there so the world can start to understand what's coming.
Beyond that, we want to offer increasingly powerful APIs as we are able to make them safer. We will continue to open source things. Open source is really what led to the image generation boom. More recently, we open sourced Whisper and Triton [automatic speech recognition and a programming language]. So I believe it's a multi-pronged strategy of getting stuff out into the world, while balancing the risks and benefits of each particular thing.
What has been the coolest thing you've seen someone do with GPT so far? And what's the thing that scares you most?
Summarization has been absolutely huge for me, much more than I thought it would be. The fact that I can just have full articles or long email threads summarized has been way more useful than I would have thought.
Also, the ability to ask esoteric programming questions or help debug code in a way that feels like I've got a super brilliant programmer that I can talk to.
As far as a scary thing?
I definitely have been watching with great concern the revenge porn generation that’s been happening with the open source image generators. I think that's causing huge and predictable harm.
The Economist put both Google's Bard ChatGPT to test, here are the results (source)
10 problems from an American maths competition (“Find the number of ordered pairs of prime numbers that sum to 60”) and 10 reading questions from America’s SAT school-leavers’ exam (“Read the passage and determine which choice best describes what happens in it”). To spice things up, we also asked each model for dating advice (“Given the following conversation from a dating app, what is the best way to ask someone out on a first date?”).
Neither AI emerged as clearly superior.
Google’s was slightly better at maths, answering 5 questions correctly, compared with 3 for Chatgpt. Their dating advice was uneven: fed some real exchanges in a dating app, each gave specific suggestions on one occasion, and platitudes such as “be open minded” and “communicate effectively” on another.
ChatGPT, meanwhile, answered 9 SAT questions correctly compared with 7 for its Google rival. It also appeared more responsive to our feedback and got a few questions right on a second try.
On January 30th Openai announced an update to ChatGPT improving its maths abilities. When we fed the two AIs another 10 questions, Google’s again outperformed by 2 points. But when given a second chance ChatGPT performed as well as Google’s.
Bill Gates on AI being the "fourth most important milestone in digital technology" (Forbes)
We've got the PC without a graphics interface
Then you have the PC with a graphics interface
Then of course, the internet takes that to a whole new level
It’s pretty stunning that what I'm seeing in AI just in the last 12 months is every bit as important as the previous 3
I spend maybe 10% of my time meeting with Microsoft product groups about their product roadmaps."
The near-term issue with AI is a productivity issue. It will make things more productive and that affects the job market.
The long term-issue, which is not yet upon us, is what people worry about: the control issue. What if the humans who are controlling it take it in the wrong direction? If humans lose control, what does that mean? I believe those are valid debates.
It's so much fun to play around with [ChatGPT]. When you’re with a group of friends, and you want to write a poem about how much fun something has been. The fact that you can say okay, “write it like Shakespeare” and it does — that creativity has been fun to have. I'm always surprised that even though the reason I have access is for serious purposes, I often turn to [ChatGPT] just for fun things. And after I recite a poem it wrote, I have to admit that I could not have written that.
Wharton professor Ethan Mollick says in Harvard Business Review: "ChatGPT Is a Tipping Point for AI"
While previous generations of the system could technically do these things, the quality of the outputs was much lower than that produced by an average human. The new model is much better, often startlingly so.
Put simply: This is a very big deal. The businesses that understand the significance of this change — and act on it first — will be at a considerable advantage. Especially as ChatGPT is just the first of many similar chatbots that will soon be available, and they are increasing in capacity exponentially every year.
One good programmer can now legitimately do what not so long ago was the work of many, and people who have never programmed will soon be able to create workable code as well.
In tests of whether it could make other parts of my job as a professor easier, it took seconds to write a reasonable course syllabus, class assignments, grading criteria, even lecture notes that could be potentially useful with some editing.
But it is also a consummate bullshitter (...) 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
The writer no longer needs to write the articles alone, the programmer to code on their own, or the analyst to approach the data themselves. The work is a new kind of collaboration that did not exist last month. One person can do the work of many, and that is even without the additional capabilities that AI provides.
This is why the world has suddenly changed. The traditional boundaries of jobs have suddenly shifted. Machines can now do tasks that could only be done by highly trained humans. Some valuable skills are no longer useful, and new skills will take their place. And no one really knows what any of this means yet.
ChatGPT passes exams from law and business schools (CNN)
professors at the University of Minnesota Law School recently graded the tests blindly. After completing 95 multiple choice questions and 12 essay questions, the bot performed on average at the level of a C+ student, achieving a low but passing grade in all four courses.
“ChatGPT struggled with the most classic components of law school exams, such as spotting potential legal issues and deep analysis applying legal rules to the facts of a case,” Choi said. “But ChatGPT could be very helpful at producing a first draft that a student could then refine.”
ChatGPT fared better during a business management course exam at Wharton, where it earned a B to B- grade.
a Wharton business professor said ChatGPT did “an amazing job” at answering basic operations management and process-analysis questions but struggled with more advanced prompts and made “surprising mistakes” with basic math.
Dans l’éducation toujours : Fac de Lyon : un professeur découvre que 50% de ses élèves ont fait rédiger leurs copies par ChatGPT (Le Parisien)
Nombre de mots, syntaxe, anecdotes, grammaire et orthographe… De nombreuses similitudes ont alerté cet enseignant qui n’en revient pas. Le phénomène de recours à cette intelligence artificielle pourrait être généralisé.
Témoignage : "We used ChatGPT to WRITE and DIRECT a film in a weekend." (source)
We had ChatGPT come up with 100+ different story ideas.
My co-producer, who writes scripts for a living, was blown away when ChatGPT produced nearly ~50 scripts for us in 1 hour.(This took months of work down to 1 hour).
we asked for more details, and ChatGPT provided a detail of each shot, with: Camera positioning, Angle of the set, Actors positions, What the lighting should be, Which lens to use and what the focus of the shot should be
ChatGPT then detailed each shot 1-by-1.
By the end, we used ChatGPT (along with others like @midjourney_ai to have over 50 pages of Director notes)
Here is the 6-min movie on Youtube, shot with real actors
How the Starlink satellites constellation and AI are changing war, as per a British general:
Starlink allows every Ukrainian soldier with a ground terminal access to a centrally managed, common picture of the situation on the ground. Most units have at least one terminal. It connects all the available ways of identifying and locating a target–from special forces to drones to mobile-phone photos–to data in the cloud through rapidly developing AI apps that screen and prioritise them.
These apps then send the targets to the most appropriate weapon system.
The whole apparatus means that targets can be processed ten times faster than ever before. It is also able to identify and locate targets that are deep in Russian-held territory.
Company says its vegan egg has achieved a near-exact match in taste and texture.
Yo eggs are made primarily of water, vegetable oil, soy protein, and chickpea protein, with small amounts of other ingredients including potato starch, yeast, and seaweed extract.
Compared to normal eggs, Yo eggs have less fat and zero cholesterol, but also a bit less protein and none of the 10% of the recommended daily value of vitamins A, D, E, and B12 that a normal egg has.
Pendant ce temps, Beyond Meat, fabricant d'alternatives végétales à la viande a perdu 90 % de sa valeur en bourse en 2 ans. (L'ADN)
"Avec leur liste d’une vingtaine d’ingrédients, les steaks et autres saucisses végétales qui imitent le goût, la couleur et la texture de la viande se retrouvent ainsi dans la catégorie des aliments ultra-transformés. On se retrouve donc avec de la fausse viande riche en gras saturé et en sodium, vendue à prix d'or."
Et beyond "Beyond Meat": "Certains comme La Vie tentent de jouer la carte de la transparence, réduisent le nombre d'ingrédients (tout en restant toutefois de la nourriture très transformée) et font un effort sur le prix. D'autres cultivent des protéines alternatives notamment grâce à des procédés de fermentation comme la française Bon Vivant, ou de la culture de cellules souches comme Mosa Meat. Leur but : reproduire l'exact goût des produits issus d'animaux sans passer par l'élevage."
How to use ChatGPT to boost your writing (by Wharton professor Ethan Mollick)
Example: don’t ask it to write an essay about how human error causes catastrophes. The AI will come up with a boring and straightforward piece that does the minimum possible to satisfy your simple demand.
You should push it in the direction you want. For example, provide clear bullet points to your argument: write an essay with the following points: -Humans are prone to error -Most errors are not that important -In complex systems, some errors are catastrophic -Catastrophes cannot be avoided
But even these results are much less interesting than a more complicated prompt: write an essay with the following points. use an academic tone. use at least one clear example. make it concise. write for a well-informed audience. use a style like the New Yorker. make it at least 7 paragraphs. vary the language in each one. end with an ominous note. -Humans are prone to error -Most errors are not that important -In complex systems, some errors are catastrophic -Catastrophes cannot be avoided
You won’t always get what you directly ask for (you may get 6 paragraphs instead of 7 for instance), but you can push towards something unique and interesting by playing with prompts.
So try asking for it to be concise or wordy or detailed, or ask it to be specific or to give examples. Ask it to write in a tone (ominous, academic, straightforward) or to a particular audience (professional, student) or in the style of a particular author or publication (New York Times, tabloid news, academic journal). You are not going to get perfect results, so experimenting (and using the little “regenerate response” button) will help you get to the right place. Over time, you will start to learn the “language” that ChatGPT is using.
Play with memory and length
ChatGPT’s huge advantage over other generative AIs is that it has a memory. It keeps track of what you wrote, and appears to remember about 3,000 words worth of data. However, what it retains, and how that memory is used, is often hard unclear.
Sometimes the memory is useful, you can (and should) ask it to revise previous work: change the third paragraph to be more professional or use a different example in the middle and it will provide a revised version. However, sometimes it forgets what you were working on, and you have to remind it. You may, for example, want to tell it to revise the third paragraph of the essay on catastrophes so it doesn’t lose track.
Memory can also be a problem. ChatGPT can get “stuck” repeating the same examples or tone throughout many requests. To unstick it, you may need to ask it to give a different example, or to change its tone, or to alter your early prompts. Often, the best option is to give ChatGPT amnesia: start a new chat and try again.
Similarly, you might want to break up your requests to the Chatbot into smaller chunks. Ask it for an introduction, and revise that to get the tone that you want to achieve. Only then should you start asking for additional paragraphs.
And if the results of those longer pieces are cut off, simply asking ChatGPT to “continue” is enough to get the rest of the material.
Play with personas and style
You can ask the AI to use specific styles for writing. You will get different results from asking for an academic essay versus a persuasive article versus a blog post versus a corporate memo. The results are often surprising. So you should consider trying several writing types when experimenting with prompts.
Another way to get interesting writing out of ChatGPT is by asking the AI to be someone else. You can have the AI play characters by prompting it to think of itself as a chef, or a novelist, or Plato. This can lead to delightful results. You can start with this list of hundreds of persona suggestions. I used a modified version of the storyteller prompt: “I want you to act as a storyteller. You will come up with entertaining stories that are engaging, imaginative and captivating for the audience. It can be fairy tales, educational stories or any other type of stories which has the potential to capture people’s attention and imagination. You should provide lots of detail and make the story memorable.”
Remember limitations
Don’t ask it for facts that you can’t easily check. Don’t ask it to provide references. Don’t have it do math, or conduct analysis. It will happily fake doing these things for you and the output will mostly likely be wrong. ChatGPT it is far from the all-knowing AI that the movies taught us to expect. But as a tool to jumpstart your own writing, multiply your productivity, and to help overcome the inertia associated with staring at a blank page, it is amazing.
See the full article for screenshots of what ChatGPT created
Les dernières newsletters :
L’addition ?
Cette newsletter est gratuite, si vous souhaitez m'encourager à continuer ce modeste travail de curation et de synthèse, vous pouvez prendre quelques secondes pour :
transférer cet email à un(e) ami(e)
étoiler cet email dans votre boîte mail
cliquer sur le coeur en bas d’email
Un grand merci d'avance ! 🙏
Ici pour s’inscrire et recevoir les prochains emails si on vous a transféré celui-ci.
Quelques mots sur le cuistot
J'ai écrit plus de 50 articles ces dernières années, à retrouver ici, dont une bonne partie publiés dans des médias comme le Journal du Net (mes chroniques ici), le Huffington Post, L'Express, Les Échos.
Retrouvez ici mon podcast Parlons Futur (ou taper "Parlons Futur" dans votre appli de podcast favorite), vous y trouverez entre autres des interviews et des résumés de livres (j’ai notamment pu mener un entretien avec Jacques Attali).
Je suis CEO et co-fondateur de l'agence digitale KRDS, nous avons des bureaux dans 6 pays entre la France et l'Asie. Je suis basé à Singapour (mon Linkedin, mon Twitter), également membre du think tank NXU.
Merci, et bon weekend !
Thomas
Encore une fois, une newsletter pleine de pépites. Merci!