AI/ML News & Innovations Hub

AI/ML news, top picks, and generated innovation digests.

★ Visit ai-karthik.com
426Sources
88150News Items
8Top Picks
439Blogs
runningLast Run

Latest AI/ML News

88150 matching items

LessWrong AI 2026-11-16 19:18 UTC Score 57.0 USR-0152-20261116-community-fo-6b6562c6

The Power of Marketing: The Case of “20 Grams of Protein” Yogurt

Epistemic status: A comparison of two yogurt labels and prices, plus a personal anecdote. AI use: Research, fact-checking, nutrition calculations, editing (grammar and light phrasing), source footnotes, and code to combine the screenshots. I've already written about marketing claims that are properties of an entire class of products . Recently, a friend of mine mentioned a yogurt with 20 g of protein, and this happened to be another example of a marketing claim that sounds more distinctive than it is. The yogurt in question is Chobani® High Protein Greek Yogurt, Lowfat Plain . Which yogurt has more protein? Chobani (left) or Great Value (right). Original product screenshots from Walmart.com, placed side by side. The 20 g claim sounds impressive, but how much yogurt is that? The label says it is per serving: Serving size 3/4 cup (190 g) Calories 140 Servings per container About 5 US serving sizes are regulated: the FDA's reference amount for yogurt is 170 g. [1] Yet these two labels use 190 g and 170 g, respectively. In this case, it is 20 g of protein per 140 calories (or 190 g) of product. The listed price at Walmart is $6.77 for 2 lb . I happened to know that lowfat Greek yogurt is high in protein in general. And indeed, the cheapest, least fancy Greek yogurt I could find , from Walmart's budget brand, has more protein per calorie (but only 17 g per serving), while costing $3.28. Item Chobani Great Value Price per 32 oz tub $6.77 $3.28 Serving size 190 g (3/4 cup) 170 g (2…

The Decoder 2026-11-16 18:52 UTC Score 57.0 AI-168-20261116-regional-ai--76ddcf0b Full article

Moonshot AI pivots Kimi models from coding to science as benchmark gains flatten

Moonshot AI is done chasing coding benchmarks, it says. The Beijing startup is pivoting its Kimi language models toward biology, chemistry, and pharma research, arguing that science demands smarter AI than law, finance, or code. The article Moonshot AI pivots Kimi models from coding to science as benchmark gains flatten appeared first on The Decoder .

CIO AI 2026-11-16 18:45 UTC Score 36.0 USR-0125-20261116-global-ai-ne-c5a38dae

Workstation pricing resets on January 1

In this issue Every price action that took effect in October, and what landed as published Lenovo’s January 1 ThinkStation increase and the GPU costs behind it The effective dates and order deadlines already published through January 2027 One planning move: get more from the devices you already own before you buy The month in brief October was the month the fall increases took hold Seven price actions took effect between October 1 and October 5: Microsoft Surface, HP Carepack services, Lenovo Top Seller, Panasonic TOUGHBOOK, and Ericsson on October 1, then Zebra and Elo on October 5. Each landed as published, with no revision received. The increases now reach well past PCs into services, rugged devices, scanners, and network hardware. New notices slowed to one, and it points at 2027. On October 19 Lenovo told partners that ThinkStation prices rise significantly on January 1, 2027, driven by NVIDIA GPU cost increases of 17-74%, with memory and SSD costs expected to follow. Orders that ship in January will be repriced. Dell has indicated a 10-20% increase in early November, the start of its fiscal fourth quarter, and Samsung’s S and Z series move November 1. Price actions in October 2026 Source: manufacturer and distribution partner communications Manufacturer Action Effective Reported scale Microsoft Surface Select Surface Pro 12th Edition and Surface Laptop 8th Edition Oct 1 Not published; as published, no revision received HP Carepack services, published SKU list Oct 1 ~7-1…

The Verge AI 2026-11-16 18:43 UTC Score 45.0 AI-016-20261116-global-ai-ne-51595064 Full article

Nvidia’s board is reportedly ‘not considering’ hiring Jensen Huang back

Nvidia's board of directors "isn't considering" having Jensen Huang return as CEO, according to a report from CNBC. A source tells the outlet the board "is focused on an investigation into Huang, and is not in discussions to bring him back." Since Huang's resignation last week, tech executives, along with current and former Nvidia employees, […]

MIT Technology Review AI 2026-11-16 18:38 UTC Score 48.0 AI-013-20261116-global-ai-ne-6352fbce

The open-source movement wants Jensen Huang back

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. On Friday, a petition to reinstate Jensen Huang as CEO of Nvidia began circulating on social media, and by Monday it listed some notable signatories, including Elon Musk and several influential Nvidia…

Kubernetes Documentation 2026-11-16 18:35 UTC Score 35.0 AI-200-20261116-developer-an-7108a573

Scheduling the removal of the kube-proxy IPVS backend

The ipvs backend for kube-proxy was an experiment in providing a service proxy implementation with better performance than the iptables backend. While it succeeded in that goal, the kernel IPVS API has turned out to be a bad match for the Kubernetes Services API, and the ipvs backend was never able to implement all of the edge cases of Kubernetes Service functionality correctly. In Kubernetes 1.29, the project introduced the nftables backend for kube-proxy, which became GA in 1.33. Part of the plan behind adding the nftables backend was the idea that it would eventually replace both the iptables and ipvs backends. Because the ipvs backend in particular is complex, not 100% feature complete, and architecturally quite different from the other two backends (and thus more work to maintain), Kubernetes SIG Network decided to officially deprecate it as of Kubernetes 1.35 . Although it currently remains available in kube-proxy (with a warning about its deprecation), we (Kubernetes SIG Network) now plan to remove it in Kubernetes 1.40. All ipvs users should therefore be planning to migrate to either the nftables or iptables backends. Why was the ipvs backend written? The ipvs backend was originally designed to fix two different kinds of performance problems in the iptables backend: First, iptables had problems with data plane latency (the speed at which clients could establish new connections to services), because kube-proxy would write a separate iptables rule to match each service…

Trump backs calls for reinstatement of Jensen Huang as Nvidia CEO
The Guardian AI 2026-11-16 18:30 UTC Score 48.0 AI-021-20261116-global-ai-ne-748f280c

Trump backs calls for reinstatement of Jensen Huang as Nvidia CEO

As tech industry pushes for return, US president says he thinks Huang ‘learned a big lesson’ Donald Trump has expressed support for the return of Jensen Huang as the chief executive of Nvidia, joining Silicon Valley heavyweights in backing the reinstatement of the AI figurehead. The US president referred to a tech industry campaign for Huang’s rehabilitation, saying a “big group” of people would like him back. Continue reading...

The iPhone Duo Is Getting a Limited Version of Face ID
WIRED AI 2026-11-16 18:26 UTC Score 54.0 AI-015-20261116-global-ai-ne-c595bae9

The iPhone Duo Is Getting a Limited Version of Face ID

Originally launching only with Touch ID, Apple’s folding phone now borrows a feature from the new HomeView and can recognize your mug, but it’s likely not as secure as Face ID.

Anthropic recognises the risks to humanity – but is its governance really up to the job? | Nils Pratley
The Guardian AI 2026-11-16 18:18 UTC Score 54.0 AI-021-20261116-global-ai-ne-bd5f279f Full article

Anthropic recognises the risks to humanity – but is its governance really up to the job? | Nils Pratley

The Claude AI developer’s unique setup is supposedly high-minded, but still represents extraordinary power concentration Boardroom governance matters, even in the world of big tech, it turns out. The business of how companies are run, investors are discovering, can be more complicated than handing the founder or superstar chief executive as much control as he wants (it’s usually a he) and waiting for brilliant returns to appear. Sometimes a few checks on power are useful, especially when a crisis strikes. This revelation is no revelation at all, of course. For about 30 years until the first generation of big US tech companies came along, the thrust of corporate governance thinking was to insist on the basic fairness of the principle of “one share, one vote”. The owners should be able to chuck out misbehaving or underperforming managers by a simple majority, in other words. Continue reading...

Why CrewAI Rebuilt Agent Memory on LanceDB, Powering 2B+ Agent Executions
LanceDB Blog 2026-11-16 18:18 UTC Score 53.0 USR-0078-20261116-ai-specialis-83e5f168 Full article

Why CrewAI Rebuilt Agent Memory on LanceDB, Powering 2B+ Agent Executions

CrewAI replaced a two-system memory stack with LanceDB, then rebuilt agent memory to resolve contradictions and gate recall on confidence. It now ships by default across 12 million monthly downloads and 2B+ agent executions.

AWS Machine Learning Blog 2026-11-16 18:16 UTC Score 68.0 AI-057-20261116-official-ai--b770d4fb Full article

How MoEngage built agentic customer engagement on Amazon Bedrock AgentCore

MoEngage built Merlin AI, a set of AI agents on Amazon Bedrock and Amazon Bedrock AgentCore that let marketing and product teams draft, analyze, and adjust campaigns from a single conversation. Learn how MoEngage turned its existing APIs into MCP tools and kept each tenant's data isolated while giving agents secure access to live platform data.

Cross Validated 2026-11-16 18:12 UTC Score 37.0 AI-113-20261116-social-media-d90f2787

Creating permutation object and running PERMANOVA for species composition in RCBD

I'm trying to determine how plant trait_a and trait_b affect insect community composition. There are two levels of trait_a and trait_b that make up four treatment groups: a1b1, a1b2, a2b1, and a2b2. For each treatment group there were five unique plant genotypes but for this example I chose three/treatment. All plant genotypes were planted in two planting blocks in a randomized complete block design. In this example I selected 2 samples from each genotype x block . Each row of the data contains species abundances/counts of insect taxa in one sample. Here is the full data frame that includes sample metadata (block, treatment, etc.): library(vegan) library(dplyr) library(tidyverse) fulldf And here is the same data without the additional sample metadata columns (formatted for vegan): specabund % unite(col = "group", sample_id, block, treatment, trait_a, trait_b, genotype, sep = "; ")%>% column_to_rownames(var="group") I want to compare my treatments across both planting blocks with PERMANOVA. I read online (Univ. Washington Multivariate Statistics Text Chapters: PERMANOVA , Restricting Permutations ) that with randomized complete block design and other complex experimental designs, permutations need to be restricted for PERMANOVA. So I created a permutation object to shuffle my data by genotype. Then I used this permutation object for PERMANOVA with the species abundance matrix as the response and trait_a * trait_b as explanatory variables: perm Is this correct? I am mostly con…

Politico Europe AI 2026-11-16 18:03 UTC Score 37.0 AI-170-20261116-regional-ai--8ad1525d

Le bras de fer entre Le Pen et les constitutionnalistes

L’initiative n’est pas habituelle : 110 professeurs de droit ont cosigné une tribune dans Le Monde, mercredi dernier, pour rappeler les règles prévues en matière révision de la Constitution. Leur message : modifier notre loi fondamentale via l’article 11, utilisé jadis par Charles de Gaulle, est irrégulier. Or, Marine Le Pen prévoit d’y recourir pour […]

Breaking down my $90,000 salary in Tokyo, Japan
Business Insider AI 2026-11-16 17:44 UTC Score 40.0 USR-0098-20261116-global-ai-ne-7f379586

Breaking down my $90,000 salary in Tokyo, Japan

Kaito left his home in LA to start a new career in Japan. The software engineer brings in $90,000 a year as a salaryman and through side gigs.

Euronews AI 2026-11-16 17:41 UTC Score 37.0 AI-164-20261116-regional-ai--af954206

Italian schools hit by new wave of bomb and poison threats

Under new security guidelines, Italian schools are no longer automatically evacuated in response to vague, mass-distributed threats, with the decision left to the discretion of law enforcement and security officials.

LessWrong AI 2026-11-16 17:39 UTC Score 50.0 USR-0152-20261116-community-fo-7fb7241f

On Friends at Frontier Labs

I am deeply uncertain about whether or not it is a good idea to get my friends to quit working at frontier AI labs. I am against people working at frontier AI labs in general, but perhaps downstream of this, all my friends in the labs have sophisticated justifications for their being there that seem plausibly true to me. More broadly, if this was not a painfully difficult question, the community wouldn't have split over it, and there are enough rationalists and EAs in the labs that it seems obvious to me that the community has split. So I don't really get my friends to try to quit their jobs, as much as try to help them regularly orient to whether or not they should — and they generally humour me by treating this as a live question. I let them know frequently that I have reservations about their work, but also that I generally trust their judgement and care about them as people. Here are my intuitions for what my friends are experiencing, while they are working at a frontier lab: They value their job. Their job is giving them silly amounts of money and validation and prestige. Their job lets them work on a really important project with really smart people in comfortable settings. Their job gives them purpose. They believe that they are doing the right thing, and that they can more or less trust the leadership, or that even if they can't trust the leadership fully, every other player is less trustworthy. They likely know that some people disagree with their choice to work at…

AI Stack Exchange 2026-11-16 17:36 UTC Score 33.0 AI-110-20261116-social-media-9eb8a06a

Can a public, white-box video verifier impose substantial per-video generation costs after adversarial optimization?

This is for a project I'm working on: an open-source CAPTCHA. A prover receives a random challenge and must submit a video of themselves performing specified actions. These might involve moving their hands in a particular sequence, manipulating objects, or performing several simultaneous actions. The submitted video is evaluated by a computer-vision pipeline consisting of multiple verifiers. Together, they produce a binary pass/fail result. Importantly, the verifiers are entirely white-box: their architectures, weights, source code and acceptance thresholds are public. They are also deterministic. The checks need not be simple deepfake detectors. They could verify geometric properties of objects, contact between objects and the performer's hands, precise finger movements, physical plausibility, deformation, multi-view consistency, etc. My question is: Can such a verifier pipeline force an attacker to incur substantial computational costs for every successfully generated video, or can an attacker eventually train a generator to satisfy the public checks with very little additional inference cost? I see two competing considerations: An attacker can inspect the public verifiers, identify weaknesses and optimize a generator against them. Verifier feedback could potentially be distilled into the generator's weights, avoiding the need to run the verifiers during every generation. On the other hand, satisfying a large conjunction of sophisticated geometric, physical and temporal co…