🚀See what happens when 50 self-driving cars go crazy on a parking ; AI spaghetti art ; fourmi chirurgienne & more
Gen AI creates videogame as you play ; robot with living skin & more
Bonjour,
Vous recevez la newsletter Parlons Futur : une fois par semaine au plus, une sélection de news, mêlant sources anglophones et francophones, résumées en bullet points sur des sujets tech 🤖, science 🔬, éco 💰, géopolitique 🌏 et défense ⚔️ pour mieux appréhender le futur 🔮.
Je m'appelle Thomas, co-fondateur de YeldaAI, qui développe des IA pour répondre au téléphone en langage humain pour les administrations et les entreprises. Plus d'infos sur moi en cliquant ici.
Et voici donc ma toute dernière sélection !
L’apéro
China has gone from being an also-ran exporter of cars to global leader in three years. (also-ran: 1. a contestant, horse, etc, failing to finish among the first three in a race. 2. an unsuccessful person; loser or nonentity.)
Apparently, Shanghai’s face recognition camera system plans to capture 26m faces per day, with scope to match against a central database of 50-60m people, and to search for features including "gender, age group, and Uyghur ethnicity." (source)
Pretty stunning AI spaghetti art... see the 20-sec video generated by AI
Nice AI-generated video of the prompt "fashion photoshoot inspired by Van Gogh”
Funny 40-sec video demo of a teleoperated humanoid robot stacking shelves with goods in a supermarket
Et une autre démo de robot humanoïde impressionnante, un modèle qui devrait entrer en masse production, say hello to Unitree G1, la démo vidéo d’1 minute
“Humanoid agent AI avatar, Price from $16K” (from website, a Chinese company)
mmm ça a l’air trop beau pour être vrai, fake à votre avis ?
Un FAIL à mourir de rire et complètement dystopique : ces voitures autonomes garées en masse sur un parking n’arrivent pas à s’accorder pour en sortir, bref un joyeux bordel.
“Tous les matins, à l'aube, sur un parking, des voitures autonomes créent toutes seules des embouteillages et une terrible cacophonie qui rendent dingues les voisins.”
“Là où un embouteillage peut être régulé avec un peu de bon sens, l’embouteillage de robots n’obéit pas à cette règle. Au petit matin, quand les premières commandes sont reçues, les taxis autonomes entament leur ballet. Ils tentent des manœuvres, reculent un peu, tournent, s’arrêtent, décident de faire le tour pour se laisser passer les uns les autres, et finissent par être tous coincés. Toutes les voitures vides se mettent alors à klaxonner, tel un troupeau d'animaux perdus.” L’ADN
La vidéo à voir absolument, un vrai côté Charlie Chaplin
L’IA générative permet même de générer un jeu vidéo en temps réel, “as you play”, dingue ! (source ici et là)
It wasn’t known this was possible until now.
Normally, game developers would define the game rules, physics and assets upfront, in order to establish how everything in the game behaves.
However, in this case, a model created a playable game just by presenting each frame one after another sequentially. It paves the way for AI-assisted game design, where developers could prototype complex games through ordinary language.
“An AI-generated re-creation of the classic computer game Doom can be played normally despite having no computer code or graphics.”
It hints at a future of infinitely variable, emergent game worlds that adapt in real time to player actions.
Voir l’exemple avec Mario et avec Doom
La résilience insoupçonnée du solaire photovoltaïque (Figaro)
Même quand la météo est maussade, le photovoltaïque est puissant : le 29 mai à midi, sous une pluie battante, RTE enregistrait une production d’électricité solaire de 7,5 gigawatts, suffisante pour recharger le parc actuel de voitures électriques françaises.
Et l’électrification est adaptée à la mobilité quotidienne. «Les automobilistes français peuvent rouler une quarantaine de kilomètres par jour. Il faudra notamment encourager, par des signaux tarifaires, les recharges en milieu de journée, quand l’électricité solaire est au plus haut de sa production et les prix au plus bas»
Yann Lecun fait une comparaison intéressante entre bébés et IAs (podcast)
"De 0 à 4 ans, la quantité d'informations absorbées par un enfant via la vue, l'ouïe, l'odorat, le toucher, soit 16,000 heures d'éveil, représente 50 fois plus qu'un LLM entraîné sur tout le texte public sur internet"
Dingue : man swallows pill containing wireless camera live on stage, streams inside of body for audience
During a fascinating TED talk, cofounder and engineer Alex Luebke showed off the company's latest invention, a tiny swallowable pill that gives physicians a live feed of the inside of the human body.
Luebke was confident enough in his startup's invention that he swallowed one of these pills live on stage, making for a unique demonstration of a potentially game-changing technology.
This pill was then remotely controlled — with a PlayStation 5 controller no less — by Mayo Clinic professor of medicine and Pillbot co-founder Vivek Kumbhari, giving audiences a fascinating live feed of the inside of Luebke's esophagus and stomach.
"So now that we're done here, PillBot will take its natural course through and out of the body," Kumbhari said. "And fortunately for Alex, he'll have no awareness of this, and he won't have to retrieve this capsule."
Dutch company ASML and Taiwanese company TSMC Can Disable Chip Machines If China Invades Taiwan. (Bloomberg)
Firms can remotely shut off advanced EUV (Extreme ultraviolet lithography) chip-making machines
The stakes are high, with around 90% of the world’s most advanced chips made in Taiwan.
What? Come on, Germany... (BBC)
82% of German companies still use fax machines on a regular basis. In many cases, fax is the go-to method for sharing medical information
Sam Altman Admits That OpenAI Doesn't Actually Understand How Its AI Works (source)
He said "We certainly have not solved interpretability,""
When pushed during the event by The Atlantic CEO Nicholas Thompson, who asked if that shouldn't be an "argument to not keep releasing new, more powerful models," Altman was seemingly baffled, countering with a half-hearted reassurance that the AIs are "generally considered safe and robust."
OpenAI Scale Ranks Progress Toward ‘Human-Level’ Problem Solving (Bloomberg)
Level 1 (where it think it's at): the kind of AI available today that can interact in conversational language with people
Level 2 (it says it's almost there): “Reasoners.” This refers to systems that can do basic problem-solving tasks as well as a human with a doctorate-level education who doesn’t have access to any tools.
Level 3: Agents,” referring to AI systems that can spend several days taking actions on a user’s behalf.
Level 4: AI that can come up with new innovations.
Level 5: AI that can do the work of an organization (same as AGI)
Former OpenAI governance researcher Daniel Kokotajlo Estimates 70%Chance That AI Will Destroy or Catastrophically Harm Humanity (!) (NYT)
According to one estimate, of the 100-odd AI experts OpenAI has hired since 2016, about half have left. (The Economist)
The result is a market where ai talent, previously hoarded at tech giants, is becoming more distributed.
La banque Citigroup prévoit que l'intelligence artificielle remplacera plus d'emplois dans le secteur financier que dans tout autre domaine (Bloomberg)
About 54% of jobs across banking have a high potential to be automated, the bank said
but...Even if AI does replace some roles across the industry, Citigroup said, the technology might not lead to a drop in headcount, as banks will have to hire more AI managers, compliance staff
Plus, new technologies haven’t always led to job cuts. In one example Citigroup offered, the number of human tellers soared between the 1970s and mid-2000s even after the introduction of automated teller machines.
L’IA générative ferait gagner 5 heures de travail par semaine à 60% des salariés, selon une étude (Figaro)
Roughly half of people who've tried ChatGPT never used it again (Ben Evans, in the WSJ)
At least 10% of research may already be co-authored by AI (The Economist)
The $700 Ai Pin, a talking lapel pin from the start-up Humane, which is funded by OpenAI’s chief executive, Sam Altman, was universally riduculed because it overheated and spat out nonsense (NYT)
SpaceX wants to build 1 Starship megarocket a day with new Starfactory (space.com)
SpaceX has added ~$100M and 21,000+ jobs to the economy in South Texas. (source)
Famous inventor Ray Kurzweil on how AI will transform the physical world (The Economist)
In all of history until November 2023, humans had discovered about 20,000 stable inorganic compounds for use across all technologies. Then, Google’s GNoME AI discovered far more, increasing that figure overnight to 421,000.
AI can rapidly sift through billions of chemistries in simulation, and is already driving innovations in both photovoltaics and batteries.
Euh.... Japanese scientists have grown living skin in a lab and attached it to a robotic face, so that it can create rudimentary facial expressions. (source)
How WW2 literally transformed the way that we see our planet (historian Adam Tooze)
Air war in particular was vastly expensive, accounting for forty percent or more of the economic war effort of the Germans, Americans or British. By the end of the war, aerospace industries were the largest sectors of the industrial economy and remade entire regions with new factories and workforces.
Out of this gigantic effort was born the technology of modern logistics, radar and sonar, a new generation of aircraft with jet propulsion, and the Manhattan Project.
Meanwhile, Germany’s vain effort to find an answer to the West’s air power gave birth to ballistic weapons programs.
By the end of the war, cameras mounted on the tips of captured German V-2 rockets fired vertically upward rather than at London or Antwerp provided the first glimpses of the outer edge of the atmosphere and the curvature of the Earth. See the video on youtube !
The war thus literally transformed the way that we see our planet as is captured in this extraordinary newsreel from 1946.
South Korea to mass produce lasers that can take out drones at $1.50 a hit (CNN)
“It is invisible and noiseless, does not require separate ammunition and can be operated only when electricity is supplied,".
Future versions could be developed to take out much bigger targets, including aircraft and ballistic missiles, which would be a potential “game changer,” according to the release.
Elephants call each other by their names (The Guardian)
Researchers used AI to analyze the calls of two herds of African savanna elephants in Kenya. They found that elephants use specific vocalizations for each individual and recognize when they are being addressed by other elephants.
Fourmi chirurgienne ! (Figaro)
Des fourmis ont été observées en train de se servir de leurs mandibules comme d’un bistouri ! Et plus précisément les fourmis charpentières de Floride (Camponotus floridanus).
Concrètement, ces insectes n’hésitent pas à amputer leurs congénères pour les sauver d’une plaie susceptible de s’infecter.
Ce comportement n’avait encore jamais été observé chez aucune autre espèce en dehors de la nôtre.
Partagez cette newsletter par Whatsapp en cliquant ici ❤️
Si vous appréciez cette première moitité de l’apéro gratuite, n’hésitez pas à prendre 3 secondes svp pour transférer cet email ou l’envoyer à ne serait-ce qu’un contact par Whatsapp en cliquant ici 🙂
Et si on vous a partagé cet email, vous pouvez cliquer ici pour vous inscrire et ne pas manquer les prochains
Je vous invite à vous abonner à la version payante pour recevoir lundi les résumés de ces articles en cliquant sur le bouton “Subscribe now” plus bas :
Vinod Khosla, Marc Andreessen And The Billionaire Battle For AI's Future (Forbes)
Billionaire investors of the internet era are now locked in a war of words and influence to determine whether AI’s future will be one of concentrated safety, or of unfettered advancement. The stakes couldn’t be higher.
Fareed Zakaria on America, The Self-Doubting Superpower, and why it may never be surpassed by China (Foreign Affairs)
How Scientists Connected 16 Mini Brains Made of Human Tissue to Create a "Living Computer"
Anduril founder Palmer Luckey shares his 3 lessons from the war in Ukraine (The Economist)
How AI is changing warfare (The Economist)
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) ou partager par whatsapp
et ici pour la version 100% en anglais
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.
Merci, et à bientôt !
Thomas