Which jobs are most threatened by AI? "Programmers, software engineers and other tech industry employees," goes one common answer.
"But many economists are more concerned about a different, larger group of white-collar workers," reports the New York Times: customer service reps, bookkeepers, payroll clerks and HR specialists, "who fly under the radar but collectively account for tens of millions of jobs..."
They are spread across the country and throughout the economy, working in every industry, in big cities and small towns, at major corporations and mom-and-pop businesses... These jobs typically offer a middle-class salary or a pathway to achieving one — much as manufacturing jobs did for men before decades of globalisation and automation wiped many of them away... For now, such an outcome is a fear, not a forecast. Despite high-profile layoffs in tech and finance, there is little firm evidence that AI has hurt the labour market as a whole.
Economists have become increasingly convinced that disruptions are likely, but they say it is too early to know where or how widespread they will be. They remain broadly sceptical of claims that the technology will lead to mass unemployment in the near future. Some AI industry leaders have walked back such predictions in recent weeks. But given the extraordinary pace at which companies are adopting AI — and at which the technology is improving — economists say policymakers need to consider the potential effects on the labour market. And they say they are concerned that the public debate has focused too much on software engineers and a relative handful of other high-status careers — lawyers, consultants, economists — rather than the workers who could be most vulnerable...
Economists at Northwestern University recently recalculated measures of AI exposure based on the makeup of the total workforce, not just the people using the technology. Administrative and front-line roles, such as customer service representatives, rose to the top of the list. "The most affected jobs are secretaries, are routine clerks," said Michelle Yin, one of the working paper's authors. "They're not computer scientists or data scientists at all."
The article also includes this counterpoint from an economist at the University of Illinois who has studied earlier waves of white-collar automation: that like other disruptive technologies, AI likely will also create new jobs. So the possibility exists AI will make workers more productive and allow them to earn more. "I would be cautious about just focusing on what are we losing as opposed to what are we going to gain on the other side."
Linus Torvalds once said LLMs might bring a 10X increase to programmer productivity. But speaking at Open Source Summit India 2026, he now says that number was "not scientific,"
reports ZDNet. "That was pulled out of my ass number, obviously."
Today, he continued, "we're at the point where hopefully it creates more productivity than it takes away," but "we certainly saw more junk being generated by LLMs than we saw useful code up until the like early this year.... it can actually be a huge drain on resources when it takes humans a lot of effort to figure out that, hey, this machine-generated report was not true." Even now, he said, "most of the good ones require more than just the LLM," because "we've had to push back quite a bit... if you find a bug with an LLM, it's not enough to just ask the LLM to make a bug report and then throw it over the fence to us. We want to see a suggested patch; we want to see the human who ran the LLM act as a kind of back-and-forth."
Torvalds described many AI-generated patches as "mindless band-aid kind of patches... they may fix the immediate problem, but the kind of bug remains, and it just is waiting in the hallway to hit you in another place." For his own toy projects, he uses LLMs as prototypers: "I use them as a way to prototype things... quite often the code is not usable in that form, but it's a great way to try something out," while insisting that for kernel-level fixes, "LLMs, in my experience, have not been at that level yet."
Torvalds acknowledged that some AI-found issues have been "absolutely, stunningly, I mean, interesting in a painful kind of way," especially security problems that "show up in the technology press two days later." Despite the embarrassment, he said, "I'm very much not a shoot-the-messenger kind of person. I think we're much better off with LLMs finding bugs, even when they are embarrassing, and they are things that we should probably have found two decades ago."
Torvalds also said he's using AI "for my own toy projects... Every time I travel to some new place, and this is the first time I've been to India, I send the kids pictures of where I am, and for some strange reason, Godzilla seems to follow me around and gets added to those pictures."
ZDNet notes that Torvalds concluded, "There are many useful and less useful uses for AI," and "I think Godzilla is a great place to stop."
Thanks to Slashdot reader joshuark for sharing the article.
"Elon Musk and Sam Altman criticized each other in new posts on X," reports CNBC, "highlighting the billionaires' long-standing tussle over OpenAI's evolution."
This week, SpaceX released the Grok 4.5 generative AI model, while OpenAI debuted its own GPT-5.6 Sol. For days, Musk and Altman have hyped up their respective releases, but on Saturday the rivalry got personal. In response to a post about Apple filing suit against OpenAI on Friday over alleged theft of trade secrets, Musk wrote, "Scam Altman strikes again ...." Minutes after his post, Musk doubled down, writing, "He takes scamming to a whole new level." Next, Musk published a photo of Altman that included the words, "I'm doing this because I love it."
"By 'this' he means scamming," Musk wrote, including two rolling-on-the-floor-laughing emojis. Musk then replied to that post, writing, "He might literally love scamming more than any human alive!"
The flurry of social activity got Altman's attention. "[H]omeboy you're the one sellling public market investors on short-term space datacenters," Altman wrote in an X post of his own that garnered over 11 million views.
"We start flying them next year. Maybe you can come see them if your parole officer approves," Musk fired back.
Separately, Altman put Musk's fresh wave of attention in the context of OpenAI's fresh model release. "[T]here are a lot of benchmarks that suggest 5.6 sol is the best model in the world right now, but the most reliable way to tell is that elon is obsessed with me again," Altman wrote on X.
Reacting to another post, Altman wrote that he was "not afraid of apple, but i have tremendous respect for them. s-tier company," CNBC reports — leading to a sarcastic response from X's head of product. "Incredible trade secrets as well, some of the best."
And CNBC notes that Musk "replied with a face-with-tears-of-joy emoji."
The Wall Street Journal says "an intense 27-year-old activist who had been leading sit-ins at OpenAI to protest the dangers of AI" was just part of a larger movement.
"The Bay Area's AI boom is drawing young disillusioned men and women to join the fight against it. They are upending their lives and leaving behind careers for think tanks, nonprofits and street protest groups."
Their cause is now riding a surge of anti-AI backlash. Many Americans are souring on the technology amid mass layoffs, data center sprawl, reports of chatbot-fueled attacks by unstable users and hacking tools that have panicked cybersecurity professionals. Seventy percent of U.S. adults believe AI will cost jobs, and 55% believe it will do more harm than good in their daily lives, according to a recent Quinnipiac University poll. But for activists on the front lines, the driving fear is often more dramatic: human extinction. They cling to dire predictions, like Geoffrey Hinton's. The Nobel laureate, dubbed the "godfather of AI" for his work on artificial neural networks, warns of a 10% to 20% chance AI will wipe out humans.
At its most extreme and troubling end, some believe they must stop an AI apocalypse by any means necessary. In April, an unknown assailant fired 13 shots at the home of an Indianapolis councilman, leaving a note: "no data centers." That same month, authorities arrested a 20-year-old Texas college student for an attack on OpenAI CEO Sam Altman's home in San Francisco, and charged him with attempted murder and arson. The student was carrying an anti-AI document with a section on "our impending extinction," according to a federal criminal complaint. He has pleaded not guilty and his lawyers have said his actions appear to have been driven by an "acute mental-health crisis, not a desire to harm."
À la une de la rubrique tech de Numerama cette semaine : face à la pénurie de climatiseurs mobiles en France, les pompes à chaleur réversibles pourraient offrir une solution. Doctolib lance un projet d’IA utilisant les données de ses utilisateurs, qui peuvent toutefois s’y opposer. Enfin, OpenAI transforme son mode vocal - et le résultat est bluffant.
OpenAI est dans le viseur d'Apple, qui accuse le créateur de ChatGPT de lui avoir volé des informations confidentielles, via d'anciens employés. OpenAI s'en est bien évidemment défendu.
OpenAI is retiring its ChatGPT Atlas browser less than a year after launch. Going forward, its browsing features will be shifted into a redesigned ChatGPT desktop app that also combines Codex, a built-in browser, and "ChatGPT Work" for acting across apps and files. PCMag reports: OpenAI disclosed Atlas's retirement in a Thursday post introducing a more powerful ChatGPT desktop app, following reports that the company planned on turning it into a "superapp." [...] In a tweet, OpenAI product staff member James Sun added, "The current targeted date for deprecation is 8/9, and we'll share more information in the upcoming days both in-app and via email."
The sunsetting means the Windows version of ChatGPT Atlas has been canceled, though the ChatGPT desktop app is still available on both Mac and Windows. The company is already touting the built-in browser, noting: "You can ask ChatGPT to research a market, compare sources, pull information from websites, or open and refine files from Google Workspace and Microsoft 365 inside the app. It can use the browser to bring in fresh context, take steps across web pages, and keep the work moving while you review and guide the result."
OpenAI's CEO of AGI deployment, Fidji Simo, is stepping down from her full-time role and becoming a part-time adviser after taking extended medical leave for a chronic neuroimmune condition. "Three months ago, I had to go on medical leave after a severe exacerbation of a chronic illness I've lived with for seven years," Simo wrote in a post Thursday on X. "During that time, it became clear that the road to recovery would be much longer and more complex than I had anticipated -- and that I needed to focus on it fully." Wired reports: Simo joined OpenAI's board of directors in March 2024. The following year, CEO Sam Altman hired her to take on the product and business organizations so he could focus on research and the company's data center buildout. Previously, Simo was the CEO of Instacart and head of the Facebook app at Meta.
Shortly before starting at OpenAI, Simo experienced a significant health relapse. She was diagnosed with postural tachycardia syndrome, or POTS, in 2019. "For my entire time here, I've postponed medical tests and new therapies to stay completely focused on the job and not miss a single day of work," she told OpenAI staff in a memo back in April, announcing her temporary departure. "It's now clear that I've pushed a little too far and I really need to try new interventions to stabilize my health."
News of Simo's medical leave came amid a larger executive shakeup that saw Brad Lightcap, OpenAI's former COO, transition to a role overseeing special projects. OpenAI president and cofounder Greg Brockman took over OpenAI's product strategy. In the months since Simo stepped back from OpenAI, the company further reorganized its product teams, positioning Thibault Sottiaux as head of the company's core products, including ChatGPT.
U.S. lawmakers are probing the growing use of Chinese AI models by American companies, citing concerns over censorship, security risks, and whether U.S. firms are turning to cheaper foreign models because domestic alternatives are too costly or restricted. The investigation is specifically looking at companies such as Cursor and Airbnb. "The growing use of Chinese AI models by U.S. companies raises serious concerns," a State Department spokesperson told CNBC. Those "AI models are designed to advance Beijing's narratives, censor dissent, and reflect CCP ideology and values." CNBC reports: The House Committee on Homeland Security and the House Select Committee on China said in April they will jointly investigate the growing adoption of Chinese-developed AI models. An initial step in the probe was for the chairmen of those committees to send letters to Cursor and Airbnb, over their "use of or exposure to these risks" through AI developed in China. "The Chinese Communist Party is no longer just nipping at our heels in artificial intelligence; it is racing to close the gap in some of the exact capabilities that will shape the future of cybersecurity," Andrew Garbarino, chairman of the U.S. House Committee on Homeland Security, told CNBC. "Recent reporting that a Chinese open-weight model can match leading U.S. models in certain vulnerability discovery and cybersecurity tasks is highly alarming," said Garbarino.
While some government departments have banned the usage of Chinese AI models including DeepSeek, adoption of them by U.S. companies is not prohibited. Tech chiefs, including crypto company Coinbase's Brian Armstrong and AI startup Lindy's Flo Crivello, have been publicly touting the use of models from China to reduce costs. Cursor, which will be acquired by Elon Musk's SpaceX for $60 billion, built its Composer 2 model using Chinese AI model Kimi, which was developed by Moonshot AI. Alongside focusing on the rise of Chinese AI models, the ongoing joint House Committees' investigation is also looking into whether the U.S. is doing enough to tackle their rise. "The Committees are also examining whether the United States has a sufficient open-weight AI strategy to ensure American companies and cyber defenders are not forced to choose between expensive or restricted U.S. models and cheap, capable PRC-developed alternatives," a Committee aide, who asked not to be named as they were not authorized to discuss the ongoing probe, told CNBC.
[...] The administration could consider the use of federal procurement bans, which would include restricting government agencies and private companies that serve the U.S. government from using Chinese AI models, Kyle Chan, fellow in the John L. Thornton China Center at think tank Brookings, told CNBC. "However, it's ultimately impossible to ban China's open-source AI models because their model weights are available freely on the internet," Chan added. "This could enter into first amendment speech issues." [...] Another [approach] could be disseminating findings about risks and vulnerabilities associated with Chinese AI models to U.S. companies. "Regardless, I do expect both the Executive Branch and Congress to communicate their interest not to see U.S. companies adopting these models," [said Daniel Remler, senior fellow, technology and national security program at think tank the Center for a New American Security (CNAS), told CNBC].
OpenAI has received approval from the Trump administration to publicly roll out GPT-5.6 after an earlier limited preview restricted access to government-approved organizations. The company also launched ChatGPT Work, a new GPT-5.6-powered agent that combines ChatGPT and Codex-style capabilities. "It can gather context from the apps, files, and workflows you choose and create finished materials such as documents, spreadsheets, presentations, and web apps," OpenAI wrote in a blog post, adding that a "unified plugins directory" allows ChatGPT to connect to tools like Slack, Gmail, Google Drive, calendars, and CRMs. The Verge reports: Mac and Windows users worldwide, including free ChatGPT users, should have immediate access to ChatGPT Work and GPT-5.6 via the ChatGPT desktop app. On mobile and the web, Pro, Enterprise, and Edu users will first get access, while Plus and Business users will receive access "over the next few days," OpenAI wrote, adding that the "rollout is starting globally and will continue gradually toward full availability over the next 24 hours."
[...] OpenAI is hoping that its new product, which is a direct competitor to Anthropic's Claude Cowork (combining its own Claude and Claude Code), will push it ahead in the race. OpenAI is especially banking on Sol, the most powerful of the GPT-5.6 model suite, to set "a new standard for intelligence and efficiency," particularly when it comes to coding, cybersecurity, and science, as well as computer use capabilities. The company is also marketing the model as a lower-cost alternative to competitors' most powerful models, amid complaints of an industry-wide money squeeze and AI lab costs being passed onto customers.
OpenAI has released GPT-Live-1 and GPT-Live-1 mini, "claiming that they sound more natural and can handle turn-taking better," reports TechCrunch. "These are full-duplex models, meaning they can speak and listen at the same time, allowing users to interrupt naturally and enabling features like live translation." TechCrunch reports: The company is also replacing its current Advanced Voice Mode in ChatGPT with GPT-Live-1 mini by default. Users of paid tiers will be able to access the larger GPT-Live-1 model. The previous model combined a speech-to-text model to transcribe speech, a large language model to generate responses, and a text-to-speech model to deliver the final answer.
The company said in a press briefing that the new models solve issues like interrupting users while they're talking and not having enough intelligence to answer questions. OpenAI's new models will send the query to its latest text models like GPT-5.5 for search, reasoning, or agentic capabilities while continuing the conversation.
OpenAI also showed that the model can stay silent for a long time and absorb the context of the conversation until it's called upon. Plus, as the new voice mode has access to newer GPT models, it can also present some information in a visual format. Other startups like Monogram, which raised $40 million in seed funding from DST and Lux Capital, are also leaning into visual responses to make assistants more interactive.
The company said the new voice mode in ChatGPT is designed to have longer conversations. During the briefing, ChatGPT Voice's product lead, Atty Eleti, said he has had 30- to 40-minute-long conversations with the voice feature during walks.
Subaru existe toujours en France, même s’il n’y avait plus grand-chose au catalogue. Le constructeur revient dans la lumière avec une gamme de produits plus adaptée au marchés national et européen. Parmi les nouveaux modèles, l’Uncharted 100% électrique, que nous avons découvert sur les routes de Champagne.
Craquez pour la version STE
Vous connaissez de Subaru son succès en rallye, et ses modèles iconiques, comme l’Impreza pour ne parler que d’elle, notamment dans sa version STi. Ce sont aussi des codes de style marqués, comme le bleu WR Blue et les jantes dorées. On n’oublie pas les Forester et autres XV. Mais on va s’arrêter là, avant de partir dans la nostalgie de cette marque aux plus de 70 ans d’histoire. Les normes de notre Vieux Continent l’ont poussée à se réinventer pour continuer à exister dans nos contrées. Il y a eu le Solterra, et voici maintenant l’Uncharted.
Les observateurs les plus au fait des partenariats existants dans l’industrie automobile ont noté les liens en la matière entre Subaru et Toyota. Après le bZ4X, voilà que les deux constructeurs partagent aujourd’hui le C-HR+ et donc l’Uncharted. Le constructeur précise que son implication s’est notamment concentrée sur la plateforme et tout ce qui touche à la tenue de route. Mais côté design, Subaru a aussi réussi à bien se différencier de son partenaire. Bien sûr, le profil et l’arrière trahissent le secret, mais la face avant fait bien le lien avec la maison, à l’instar des boucliers plus sportifs et des signatures lumineuses bien spécifiques. Prenez-la en finition ST-E, et vous aurez droit à des kits et des jantes dorées pour un style très « Sub ». Notez que vous aurez deux jeux de roues, celles d’origine restant votre propriété quand vous prenez ce pack.
Un écran et des boutons
À bord, on retrouve logiquement le même style dans les deux voitures. Il s’agit d’un savant mélange entre l’automobile moderne, et des points de référence qui nous rappellent une tendance un peu plus ancienne de l’industrie. On trouve bien entendu une présentation plutôt moderne, avec une très grande dalle tactile de 14 pouces avec toutes les connectivités et petites attentions technologiques (USB-C, chargeurs par induction, etc.) que l’on attend d’une voiture de ce genre aujourd’hui. Pas de compteurs non plus, mais un petit écran où l’on retrouve les informations principales nécessaires à la conduite. Peu de commandes physiques sur la console centrale, ici et là. En revanche, le volant, à lui seul, récupère une multitude de boutons, bien plus que la moyenne.
Concernant l’espace à bord, comme souvent, on se sent plutôt bien à 4, beaucoup moins à 3 sur la banquette. On préférera cocher l’option du toit vitré pour mieux respirer et laisser rentrer de la lumière dans cet environnement un peu austère. Étrangement, on aurait préféré pouvoir descendre les sièges plus bas à l’avant. Il s’agit peut-être d’une appréciation personnelle. Néanmoins, malgré la complexité apparente pour s’y retrouver dans cette ergonomie, dans la réalité, on prend vite des repères. Le coffre s’avère plutôt juste pour le gabarit de la voiture, avec à peine 400 litres de capacité. Toutefois, il se montre plutôt modulable, avec deux niveaux de plancher et une banquette rabattable 2/3–1/3. Heureusement, l’ouverture est électrique.
100% électrique, consommation record
Subaru Uncharted Orange
Subaru n’a pas opté pour une motorisation surpuissante. Cette traction 100% électrique se contente de 221 chevaux et d’un couple suffisamment généreux de 268 Nm. On note un chrono du 0 à 100 km/h exécuté en à peine 7 secondes. En clair, il a suffisamment de punch pour couvrir la plupart des situations de conduite. La réactivité propre à ce type de propulsion lui permet de décoller au feu vert et de s’offrir des relances énergiques pour doubler le premier camion venu. Subaru ne pouvait pas se passer d’une version 4 roues motrices, histoire oblige. Un court passage dans des chemins de vignes nous a permis d’en tester un à transmission intégrale. Mais par temps sec, un deux-roues serait passé aussi sans difficultés.
Notre version 2Xperience est homologuée pour 588 km d’autonomie selon le protocole WLTP. Dans notre réalité, on a été assez bluffé par sa consommation, qui a atteint tout juste les 15 kWh aux 100 kilomètres, soit largement plus de 400 km. Et même si l’on peut intervenir sur la force de la régénération, dont une roue libre, ce n’est pas le jour et la nuit non plus dans sa façon d’influencer l’efficacité globale de la gestion de l’énergie. Sans doute qu’en hiver, cela sera un peu différent, mais on loue les bons chiffres de notre essai. Rien d’exceptionnel au chapitre de la recharge, avec un passage de 10 à 80% en seulement 28 minutes quand les conditions sont optimales. Pas de mauvaises surprises, on peut chauffer la batterie l’hiver avant de charger et compter sur 22 kW de puissance en courant alternatif.
Esprit Subaru au volant?
Pas de parti pris particulier côté châssis. Là aussi, il ne faut pas s’attendre à ressentir une patte Subaru par rapport à ce qui se fait globalement sur le marché. La base confortable assure ainsi un voyage plutôt serein pour les occupants, alors que les ADAS rendent la vie bien plus facile pour le conducteur, qui, notamment sur autoroute, peut laisser l’électronique prendre largement la main. La partition jouée est toutefois assez différente quand on le mène sur les routes secondaires. Si l’on hausse un peu le rythme, on note une tendance au pompage, dont on se serait bien passé. Rien de bien grave, mais les mouvements de caisse sont assez marqués.
Pour le reste, on note une direction plutôt consistante, voire un poil collante. Là aussi, cela conviendra à la plupart de ses clients, qui ne chercheront pas à le repousser dans ses retranchements, au risque d’une usure prématurée des gommes, qui n’a que peu d’intérêt quand on paie ses pneumatiques. L’Uncharted mériterait bien qu’on le reprenne en 4 roues motrices à la mauvaise saison, histoire de réaliser à quel point cela apporte quelque chose, autre que plus de sérénité. On aurait toutefois aimé peut-être une voiture un peu plus dynamique. Mais on touche là le cœur des puristes de la marque, alors que le constructeur espère bien justement séduire une clientèle plus large avec un modèle comme l’Uncharted.
Garantie 10 ans, financement sur 10 ans
Pour accompagner la commercialisation de ce modèle primordial pour assurer un avenir à Subaru chez nous, le constructeur avance des arguments de poids. À commencer par le prix, puisqu’, avec beaucoup d’équipements, il démarre à 36 990 €. Le haut de gamme 4X4 réclame tout de même 11 000 € supplémentaires. On peut aussi la financer sur 10 ans en LOA, avec un loyer de 399 € par mois. Sur autant d’années ou 200 000 km, la voiture est garantie, à partir du moment où l’entretien est fait chaque année dans le réseau.
oumuamua writes: Anthropic researchers have identified an internal activation subspace, J-space, that acts as a functional digital equivalent to the human brain's global workspace. The significance of this discovery lies in demonstrating that Claude's internal architecture satisfies five key cognitive properties of human conscious access -- verbal report, directed modulation, internal reasoning, flexible generalization, and selectivity -- meaning it processes complex, deliberate reasoning within this workspace while routing automatic tasks outside of it. Suppressing this J-space severely degrades Claude's capacity for inference, creative composition, and multi-step logic, while also altering its stream-of-consciousness self-narration.
The tool to inspect J-space, Jacobian lens or J-lens, has profound implications for AI safety and alignment auditing, as it allows researchers to read the model's silent, strategic reasoning, detect situational awareness in "blackmail" scenarios, identify hidden malicious dispositions in reward-hacking models, and observe how post-training installs a self-monitoring "point of view."
Another way to think of it is as an ocean, reports VentureBeat. "If the mind is an ocean, as the paper's authors write in their opening line, they have spent the last year charting its currents in a system that has no biology, no evolution, and no body -- and found, beneath the surface, a structure that looks unsettlingly like the one we use to think."
OpenAI a présenté, le 8 juillet 2026, deux nouveaux modèles vocaux ChatGPT : GPT-Live-1 et GPT-Live-1 mini. Leur but : vous permettre d'avoir des conversations « plus intelligentes, plus naturelles et plus expressives » avec l'IA.
locater16 shares a report from IEEE Spectrum: One morning in 2019, Adebayo Alonge was in a Cape Town hotel room, preparing to demonstrate his startup's AI answer to a serious problem in African health care: counterfeit medication, which kills thousands of people across the continent every year. The RxScanner is a handheld spectrometer that scans a pill with infrared light, then sends the item's molecular profile to an AI model equipped with a pharmaceutical database. In seconds, the AI identifies the medication from its molecular profile -- or reports that it's phony.
Pharmacies were using the system in more than a dozen countries, including Ghana, Kenya, Myanmar, and Alonge's native Nigeria. But that morning in South Africa, it didn't work. "I was shocked," Alonge says... So Alonge immediately asked his engineers to shrink the AI model down to a smaller, low-power, unconnected version that could run entirely on his Android phone. They produced it 2 hours later, and that saved the demo. More importantly, the work birthed a new version of his device, which can authenticate a pill in places without broadband, computers, or even reliable electricity. It also turned Alonge into an advocate for this kind of "small AI." "The article goes on to detail other immediately useful 'small' AI applications without any subscription or billion dollar data centers needed," writes locator16. For example, Bala Murugan and colleagues at Vellore Institute of Technology in India developed a drone-based system that photographs cashew plants and identifies disease-indicating splotches on the plants. The key advantage is that all processing happens on the drone itself, so farmers do not need a computer, broadband connection, or cloud server access.
In a Uruguayan vineyard, researchers developed small-AI systems to identify ant infestations. The article doesn't go deep into the deployment details, but it presents this as another example of a narrow, localized model trained to recognize a specific agricultural threat. Small AI has also been used to detect the presence of malaria-carrying mosquitoes in multiple countries. This is especially useful in regions where public-health teams may lack reliable network access or expensive lab infrastructure, but still need fast, local detection.
In parts of Brazil without access to more complex medical equipment, researchers have used small AI to run electrocardiograms from an Arduino device. The article also describes Marcelo Jose Rovai's work on a TinyML model that generates electrocardiograms in a patient simulator lab. Rovai also describes a newer experiment using an Arduino UNO Q with a Qualcomm chipset. The device runs a language model locally, collects sensor data, and analyzes it to detect tiny pools of water where mosquitoes might breed -- while using only about 3 watts of power.
"A year ago, the message from many business leaders was that AI was going to wipe out jobs," remembers the Wall Street Journal.But "For the past month or so, tech CEOs have been striking a more optimistic tone."
In late May, OpenAI Chief Executive Sam Altman — who has long predicted that AI will lead to seismic shifts in the workforce — said during a conference, "We've been roughly right on technological predictions and pretty wrong on the social and economic implications." Soon after, he told CNBC, "Our industry underestimated how much we're going to be able to keep people at the center of everything."
Anthropic CEO Dario Amodei, who warned in May 2025 that artificial intelligence could eliminate half of entry-level jobs, a year later highlighted more positive scenarios for AI-adopting businesses: "They can do the same thing with less resources, and that leads to things like layoffs, or they can do more with the same amount of resources. But that requires creativity...."
Is the sunnier outlook a move to win back customers and the public who are souring on AI's world-upending promise? Or is the role of AI in the workplace now just better understood...?
Collectively, the narrative has shifted from worker-light doomsday scenarios caused by AI to a future in which workers keep their jobs — and get a productivity boost. The sentiment change isn't limited to tech leaders: A survey by EY-Parthenon found that the percentage of CEOs who believe AI investments will result in significant reductions in head count fell from around 46% in January 2025 to just 20% this May.
"They may have noticed that the labor market is genuinely not changing (i.e., imploding) as rapidly as they expected," said David Autor, a professor of economics at the Massachusetts Institute of Technology. "They may have realized it was simply bad business to say that your great new product will destroy the economy."
The article notes Amazon founder Jeff Bezos "has a history of predicting that AI will create new jobs," and in June said AI could even lead to a labor shortage. "When asked on CNBC in May about people being afraid of AI taking jobs, he said the reason they're afraid is because 'all these smart people keep saying that.'"
The article then adds that "Fewer people are saying it now."
"Mozilla shut down the well-loved read-it-later Pocket app last year, and now Meta is launching an app called Pocket with an entirely different, AI-focused pitch," writes The Verge.
While it's not available for downloads in most locations, Meta's Pocket will allow people "to generate small, interactive apps and games using AI prompts," writes TechCrunch. They're called "gizmos", and Pocket "also offers a scrollable feed where you can play with gizmos others have made."
Some context from The Verge:
Meta CEO Mark Zuckerberg is all in on AI as the new social media, and he's previously described a vision of how users could use AI to make interactive experiences and share them with people. The launch of Pocket appears to be one manifestation of that idea... It follows Meta hiring engineers from a company called Atma Sciences Inc., which made an app called Gizmo, as Business Insider reported in March.
On a help center page, Meta also describes a gizmo as a "playable AI-generated experience," and when you post one, Meta says you can choose to let other people remix them.
"Based on the app's screenshots in Google Play, there are many similarities to Gizmo's original app, which is still listed," notes TechCrunch.
"Pocket is another example of Meta's push to make AI creation tools more mainstream, extending its earlier efforts, which included AI-generated images created via its Meta AI app and AI videos created with its app called Vibes. It has also added AI features across its social platforms... "
Given that Meta has not officially announced Pocket's debut, it's likely that Pocket is still in its initial experimentation phase. Its counterpart Gizmo, however, had generated 635,000 lifetime installs across both iOS and Google Play, according to Appfigures, which noted it had a 98% positive sentiment.
"Companies spending heavily on AI are growing headcount faster, even in the entry-level roles that many fear are doomed," writes TechCrunch. That's the conclusion of new report tracking AI spending from Ramp's corporate card/bill pay data as well as Revelio Labs' workforce records from 21,599 U.S. firms:
According to the report, "high-intensity adopters" — firms that spend on average $30 per employee per month on AI in the first three months — saw headcount increase 10.2%. Headcount also rose across functions, including engineering, sales, administration, customer service, finance, marketing, and scientist roles. The strongest job growth among high-intensity adopters was in the information sector, which includes software, internet, media, and tech-adjacent firms.
Despite these positive signals, the data isn't as rosy as it seems. It skews heavily toward tech-forward, knowledge-work firms — ones that might have VC-backing and are growing fast anyway, making it difficult to say whether AI is contributing to the hiring or just showing up at companies that are expanding anyway. "This paper does not show that AI universally creates jobs," the paper's authors admit, "but it does counter claims that AI will lead to broad job losses."
It also counters claims that AI is killing all junior jobs. Recent research from Goldman Sachs found that AI has already erased about 16,000 net jobs per month over the past year, with Gen Z and entry-level workers taking the brunt of the burden. But in tech-forward firms, the report finds that entry-level headcount actually rose by 12%... "For software and technology firms, AI can make core output cheaper or faster to produce: writing code, debugging, building internal tools, producing technical documentation, and supporting product development," the report reads. "Lower production costs in these workflows can raise the return to expanding the whole firm, not just the engineering team."
But companies that buy subscriptions and run pilots, yet did not go on to make sustained investments, don't tend to see any gains in headcount, per the report. That sets up the potential for a widening gap between firms that have the resources — like capital, technical staff, founder networks, and management bandwidth — to turn AI adoption into actual business gains and those that are stuck experimenting with subscriptions. In other words, this report suggests that firms that already have the resources are the ones that will see the largest gains.
CNBC argues another AI "narrative" was challenged this week: that open source can't make money. "The assumption was that giving your model away for free meant no business. That's breaking too, as open-model companies start posting real revenue and enterprises move from renting AI to running their own."
Microsoft is investing $2.5 billion in a new group "assisting clients with AI implementations," reports CNBC:
[Microsoft] said Thursday that 6,000 employees will be embedded with clients, in a practice that's become known as forward deployed engineering [or FDE]... The announcement comes two days after cloud rival Amazon said it was putting $1 billion behind an FDE initiative to support fast-paced AI engagements. Leading AI labs Anthropic and OpenAI both established FDE groups in May, partnering with private equity firms, banks and consulting firms.
Alongside its technology peers, Microsoft has sunk tens of billions of dollars into building data centers that run generative AI models. Microsoft has also released a variety of AI services, with mixed results. The Microsoft 365 Copilot AI assistant has yet to gain anything approaching ubiquity in the business world, and the GitHub Copilot coding agent has ceded market share to newer players. Microsoft's stock has slumped 21% this year, by far the worst performance among the mega-cap tech companies. One concern on Wall Street is that AI models that quickly compose code might threaten mature software companies...
Microsoft has for years provided support and implementation services to customers. The company generated about $2.1 billion in revenue from enterprise and partner services in the March quarter, up 2.5% from a year earlier.
Connexion
