Zendoric

Zendoric

AI analysis and perspective, every day.

Updated: July 21, 2026
🎙️ Listen to today's podcast (in Spanish)

🔥 In focus today

Apple expands its legal offensive against OpenAI, targeting dozens of its former employees

🕒 Published on Zendoric: July 21, 2026 · 00:20

Reader notice: the original article (Financial Times, via Financial Post) is mostly behind a paywall. The page shows subscription notices, but also reveals a considerable amount of body content, which is what is summarized here.

China bans AI 'boyfriends': new national regulation limits companion chatbots

🕒 Published on Zendoric: July 21, 2026 · 00:20

China has implemented, starting July 15, the first nationwide regulation aimed at curbing the emotional dependency many users are developing toward artificial intelligence chatbots designed as virtual partners, friends or confidants.

Apple's Trade-Secret Suit Against OpenAI Threatens the Hardware Bet and the IPO Pitch

🕒 Published on Zendoric: July 21, 2026 · 00:20

Apple has sued OpenAI over trade secrets, accusing it of a pattern of misconduct to pull confidential information from current and former Apple employees — and naming hardware chief Tang Tan directly. The timing collides with OpenAI's rumored smart-speaker device and its confidential IPO filing, raising the question of whether litigation, not competition, becomes the real obstacle to its hardware ambitions.

AI Voice Cloning Fuels Nearly $900M in US Scam Losses, FBI Data Shows — and Georgia Isn't Spared

🕒 Published on Zendoric: July 21, 2026 · 00:20

The FBI's 2025 Internet Crime Report puts AI-enabled scam losses at almost $893 million nationwide, with Florida and Georgia together reporting over $50 million in AI-related fraud. From a cloned attorney's voice targeting immigrants to an $800,000 crypto scam against a Cobb County couple, the cases show how generative AI is making old scams unrecognizably convincing — and why verification, not speed, is becoming the only real defense.

📰 More AI analysis

AI enters the senior shopping basket: 1 in 5 people over 50 already use it to compare prices

🕒 Published on Zendoric: July 21, 2026 · 00:20

19% of people over 50 already use artificial intelligence to compare products before buying, according to Cetelem's Consumo Europa 2026 report. But 79% still trust brick-and-mortar stores, and a quarter of those over 60 have never shopped online: the senior demographic isn't uniform, it's splitting in two.

A Greenwich school cuts classes to two hours a day with AI: the rest, life skills

🕒 Published on Zendoric: July 21, 2026 · 00:20

Alpha School, the private school network that replaces much of traditional teaching with adaptive software, opens its first Greenwich (Connecticut) campus this fall for up to 100 students from kindergarten to high school. It promises results in the top 1-2% of the US with just two daily hours of academics; the rest, workshops on finance, public speaking and business. The proposal divides parents and students in the district.

Sheinbaum proposes a census to regulate AI and social media around children, but without a roadmap

🕒 Published on Zendoric: July 21, 2026 · 00:20

Mexico's president opened the debate on regulating social media and artificial intelligence with a focus on protecting children, and proposed building a prior census. The announcement marks the start of a discussion, not a policy: timelines, responsible parties, and a definition of what would be counted are still missing.

Tempus AI pays $1.5 billion for Personalis: it buys the cancer test it already sold exclusively

🕒 Published on Zendoric: July 21, 2026 · 00:20

Tempus AI acquires Personalis for $1.5 billion in stock, three years after becoming its commercial partner. The goal: to control end-to-end a residual cancer detection test whose sales are already growing 33% quarterly in a barely tapped $20 billion market.

$1.5 billion: judge approves the largest copyright settlement of the AI era with Anthropic

🕒 Published on Zendoric: July 21, 2026 · 00:20

A federal judge in San Francisco gave final approval to the $1.5 billion settlement between Anthropic and a group of authors, the highest known payout in a US copyright lawsuit. The case sets a precedent that will shape the whole industry: training on books can be fair use, but hoarding pirated copies is not.

Google prepares a chip up to 10 times more efficient for Gemini and calms Wall Street ahead of earnings

🕒 Published on Zendoric: July 21, 2026 · 00:20

Alphabet is developing 'Frozen v2', a new server chip that, according to The Information, could multiply the energy efficiency of its Gemini models by six to ten times. It will arrive in 2028, Google neither confirms nor denies it, and its stock already rose 3% on the rumor alone.

Medellín teaches AI to teenagers through urban music: literacy, not just the model, is the race that matters

🕒 Published on Zendoric: July 21, 2026 · 00:20

Medellín's Department of Education is launching the Urban Labs call for eighth to eleventh grade students to blend artificial intelligence and urban music in projects about their own neighborhood. It's a modest program, but it points to a question that will define the next decade: who gets early access to AI and who doesn't.

Sally, the $57,590 robot arriving at a rural high school in New York (and not here to replace anyone)

🕒 Published on Zendoric: July 21, 2026 · 00:20

A small school district in western New York, on Seneca Nation land, will introduce a humanoid robot from Realbotix into its AI and robotics classes this fall. The superintendent himself rushes to clarify the obvious: it doesn't replace any teacher. The case, modest in scale, is a good gauge of how AI is really entering the classroom.

AI against cancer: Harvard's model hits 94% accuracy and China tests a vaccine that hasn't yet reached human trials

🕒 Published on Zendoric: July 21, 2026 · 00:20

CHIEF, a model from Harvard Medical School published in Nature, diagnoses cancer with 94% accuracy and outperforms other AI systems by more than 35%. Meanwhile, China is advancing an AI-guided personalized cancer vaccine that, for now, has only worked in animals.

MCP, the 'plumbing' connecting AI to your apps, fixes a design flaw to scale to millions of users

🕒 Published on Zendoric: July 21, 2026 · 00:20

The Model Context Protocol (MCP), the standard that lets models like Claude connect to Gmail, Slack or internal databases, rolls out a technical update next week: it changes how it manages sessions to stop clashing with large companies' load balancers. It's not a flashy update, but it could be the one that truly moves the needle on agentic AI in production.

China bans AI 'boyfriends' by law and triggers a wave of breakups: behind it lies birth rates, not just wellbeing

🕒 Published on Zendoric: July 21, 2026 · 00:20

Beijing has banned, effective July 15, AI companion apps from generating 'emotional dependency or addiction,' and several tech companies have directly shut down their virtual characters. Thousands of users describe the loss as genuine grief. The underlying reason, according to the Wall Street Journal, is China's birth rate at historic lows.

A medical framework puts the patient ahead of savings in the hospital AI race

🕒 Published on Zendoric: July 21, 2026 · 00:20

Two doctors from UVA Health and Clemson publish a framework—'Total Mission Value'—for hospitals to evaluate AI by its clinical and human impact, not just price. It's an attempt to bring judgment to an adoption process that's advancing faster than the tools to assess it.

Is 'graph engineering' real? From loop engineering to the debate on agent graphs

🕒 Published on Zendoric: July 21, 2026 · 00:20

This edition's FOD editorial analyzes a very recent phenomenon in the public conversation about agentic AI: the emergence of the term 'graph engineering' barely six weeks after 'loop engineering' became popular, and how within days the community is already declaring the latter dead...

AI is more prone than humans to develop its own biases when hiring staff

🕒 Published on Zendoric: July 21, 2026 · 00:20

A new study by researchers from Princeton University and the University of Chicago, presented in a paper at ICML (Seoul, July), shows that large language models (LLMs) not only inherit biases from their training data but can also generate their own stereotypes from experience...

Sabotage of weather data: a growing risk amid prediction markets and the AI boom

🕒 Published on Zendoric: July 21, 2026 · 00:20

Weather forecasts are not a trivial matter: airlines, grid operators and farmers make daily decisions based on them, from which crop variety to plant to where to build solar or wind farms and how to price wholesale electricity.

SpaceX negotiates selling the Pentagon AI computing capacity worth billions of dollars

🕒 Published on Zendoric: July 21, 2026 · 00:20

This article comes from The Wall Street Journal and is behind a paywall: the downloaded content is only a brief snippet (teaser), so the summary below is strictly limited to that data, without filling gaps with assumptions.

Trump Media asked for $100,000 a month for early access to Trump's posts

🕒 Published on Zendoric: July 21, 2026 · 00:20

The content downloaded from this Financial Times link corresponds entirely to a paywall: the page does not offer the body of the article, only the site's navigation menu, listings of other featured news, and the various subscription options (trial for €1, Standard, Premium plans, and…

ICE shared Medicaid data it shouldn't have had with Palantir, new court revelations show

🕒 Published on Zendoric: July 21, 2026 · 00:20

A federal court case has revealed that the US Immigration and Customs Enforcement (ICE) shared with data analytics company Palantir Medicaid information that health authorities themselves had improperly handed over to it.

Apple dethrones Nvidia as the world's most valuable company amid a shift in AI bets

🕒 Published on Zendoric: July 21, 2026 · 00:20

Apple surpassed Nvidia on July 17, 2026 to become the world's most valuable company, on a day when its stock stayed stable while Nvidia's shares fell 3.5%.

Armed robots: Washington opens the door to autonomous weapons powered by Silicon Valley

🕒 Published on Zendoric: July 21, 2026 · 00:20

The content downloaded from this Washington Post article corresponds only to a fragment before the paywall, so the analysis that follows is necessarily brief and limited to what appears literally in that fragment.

Scientists work on lab-grown teeth to replace fillings and implants

🕒 Published on Zendoric: July 21, 2026 · 00:20

Surveys cited in the article suggest that up to 70% of adults fear going to the dentist, leading many people to avoid check-ups and routine care.

Kimi K3 leads programming benchmarks (unverifiable content)

🕒 Published on Zendoric: July 21, 2026 · 00:20

It was not possible to access the actual content of this link. All the download returned was YouTube's standard cookie notice and consent screen, with no transcript, description or data about the video itself.

Alphabet shares drop after reported delay in its Gemini 3.5 Pro AI model

🕒 Published on Zendoric: July 21, 2026 · 00:20

Alphabet shares fell 4% on Thursday after Bloomberg reported, citing sources familiar with the matter, that the launch of the company's flagship artificial intelligence model, Gemini 3.5 Pro, is months behind the originally planned schedule.

Netflix reveals about 300 productions used generative AI in 2026, according to its earnings report

🕒 Published on Zendoric: July 21, 2026 · 00:20

Netflix confirmed in its investor letter for second-quarter 2026 results that around 300 shows in its catalog used generative artificial intelligence at some point in their production process this year.

How to successfully scale AI programs in the enterprise

🕒 Published on Zendoric: July 21, 2026 · 00:20

McKinsey senior partner Brooke Weddle explores, in this installment of the "Rewired takes" series, how companies are successfully scaling their artificial intelligence programs, focusing on the practical "people lessons" needed to get there.