AI/ML News & Innovations Hub

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

★ Visit ai-karthik.com
426Sources
90067News Items
8Top Picks
445Blogs
successLast Run

Latest AI/ML News

90067 matching items

Can you pass our science quiz set by Nobel laureates?
New Scientist AI 2026-11-13 13:00 UTC Score 34.0 AI-027-20261113-global-ai-ne-5703bec6

Can you pass our science quiz set by Nobel laureates?

Test your knowledge of Nobel-prizewinning discoveries in microbiology, genomics and cosmology with questions set by the laureates themselves: Barry Marshall, Jennifer Doudna and Saul Perlmutter

Apple should have made the MacBook logo RGB
The Verge AI 2026-11-13 13:00 UTC Score 49.0 AI-016-20261113-global-ai-ne-715096a9 Full article

Apple should have made the MacBook logo RGB

When Apple CEO John Ternus announced the first touchscreen Mac last month, the biggest reaction in the room wasn't for the PROLED MacBook Pro's OLED touchscreen or the thinner and lighter chassis; it was for the illuminated lid logo. It's a throwback to the light-up logos on the old MacBooks, which Apple started phasing out […]

Trump threatens new tariffs on Mexican goods
Semafor Technology 2026-11-13 12:40 UTC Score 49.0 USR-0094-20261113-global-ai-ne-517b83d4

Trump threatens new tariffs on Mexican goods

The US president threatened to double tariffs on Mexican goods if the US’ southern neighbour — and top trading partner — doesn’t stop a migrant caravan heading for the border.

InfoWorld AI 2026-11-13 12:37 UTC Score 34.0 USR-0126-20261113-global-ai-ne-5a34ea18 Full article

Google adds multi-model Orchestrator, Gemini 4 Hydrogen, to Antigravity

Google is adding a new feature, Orchestrator, to Antigravity, its AI-powered development environment, to help enterprises improve software development efficiency by automatically coordinating multiple Gemini models, including the newly introduced Gemini 4 Hydrogen. The new feature, which allows developers to delegate planning, coding, and review tasks to different models within a single coding session, combines a planning model, such as Gemini 4 Carbon or Argon, with Hydrogen, which the hyperscaler describes as a lighter execution model. While the planner models explore the codebase, identify dependencies, and develop an implementation plan, Hydrogen handles execution, including editing files, writing code, and running builds and tests, the hyperscaler wrote in a blog post . That execution differs from conventional model routing, which typically selects a model based on a prompt and assigns it the entire task. Orchestrator, instead, transfers control between the planning and execution models as work progresses, with the planning model first exploring the codebase, identifying dependencies, resolving ambiguities, and preparing an implementation plan before handing control to Hydrogen for execution, the hyperscaler explained in the feature’s documentation . Hydrogen hands control back to the planner if it encounters problems, such as repeated test failures or unexpected behavior, allowing the latter to reassess the execution and revise the plan before Hydrogen resumes implemen…

I'm leaving my PhD to pursue AI Safety work
LessWrong AI 2026-11-13 12:32 UTC Score 71.0 USR-0152-20261113-community-fo-2f7c3948 Full article

I'm leaving my PhD to pursue AI Safety work

Yesterday, I told my advisors that I would be dropping out of my PhD program at the University of Michigan's School of Information. I was nervous about how to break the news to them. There was already some tension in our weekly meetings: I often voiced my dissatisfaction with the pace of academia, my interests had been drifting away from theirs, and, most of all, we disagreed on the severity of the risks AI posed. When I walked into the room, they told me something I would have loved to hear months earlier: “We think it’s time to get you started on some research.” All I could say was, “I think it’s a little too late for that.” For months, I had been proposing projects that bridged my interests with theirs, such as running studies that treated humans and LLMs as equal subjects, but they were usually turned down for not being “exciting enough”. My advisors wanted me to start with small, unpublished surveys that would ‘help me ask the right questions’ and “develop research taste” in Human-Computer Interaction before I could propose a full study. I understood why they wanted me to gain experience first, but I felt held back after every meeting, and grew restless alongside my ambition and sense of urgency. “I’ve been doing machine learning research independently,” I told them. “I discovered some really interesting stuff about how to interpret neurons inside of LLMs. It’s garnered some attention, and some people are even open to funding my work. I’ll be withdrawing and moving to S…

Jensen Huang Resigns After Having Another Affair In Addition to the Chinese Spy
LessWrong AI 2026-11-13 12:30 UTC Score 66.0 USR-0152-20261113-community-fo-77a3110a Full article

Jensen Huang Resigns After Having Another Affair In Addition to the Chinese Spy

Jensen Huang has resigned as CEO of Nvidia and left its board. CFO Colette Kress will be interim CEO . Here is a quick profile of Kress . Here is a look at some of the potential long term successors . There were a lot of efforts to bury the fact that Jensen Huang had an affair with a Chinese spy, and then even to bury his second affair as well, and to get everyone to ignore the obvious implications of this. Then the second affair was revealed with executive Kimberly Powell, who directly reports to him. Even then, many doubled down on defending Jensen Huang as CEO of Nvidia. The Nvidia board disagreed, and was preparing to oust CEO Jensen Huang prior to his resignation. This creates a tricky situation for Nvidia, given that Jensen Huang had over 50 direct reports and did not believe in succession planning, and also for the White House . It also creates a tricky situation for many in Silicon Valley, who have invested heavily in their relationships with Jensen Huang and in the idea that founders are as Gods. This post covers all related events since my initial post . If you haven’t seen it, this short AI video is obligatory , as is this one . The Daily Show has coverage of stage one. They weren’t being all that subtle, huh? Odd Lots had an essay on the Megaspeed scandal . As Tracy Alloway says, one simple takeaway is that Nvidia could have easily sold all these chips to Americans. Matt Levine has additional coverage . Derek Thompson summarizes this ‘insane scandal.’ And of cour…

Medianama AI 2026-11-13 12:24 UTC Score 44.0 USR-0211-20261113-regional-new-dd9fcf6a

Turtlemint flags lower income for agents under IRDAI proposed commission curbs, pushes beyond insurance

Turtlemint says IRDAI's proposed commission caps could reduce monthly PoSP earnings by Rs 1,000 more than those of insurance agents selling the same number of policies. The post Turtlemint flags lower income for agents under IRDAI proposed commission curbs, pushes beyond insurance appeared first on MEDIANAMA .

Hungary’s MOL seeks EU funding as CEO warns Russian oil phase-out could trigger social unrest
Euronews AI 2026-11-13 12:16 UTC Score 42.0 AI-164-20261113-regional-ai--1734f71e Full article

Hungary’s MOL seeks EU funding as CEO warns Russian oil phase-out could trigger social unrest

MOL says it could phase out Russian oil imports by the end of 2027, but chief Bacsa warns that higher energy costs could fuel political instability in Hungary and Slovakia. He is calling for EU funding to expand alternative pipeline routes, as Brussels keeps focus on quitting Russian energy.

We want to run AI safety events for you!
LessWrong AI 2026-11-13 12:13 UTC Score 60.0 USR-0152-20261113-community-fo-4bcfe7e8 Full article

We want to run AI safety events for you!

TL;DR: We’re four undergraduate organizers who want to build an Events-as-a-Service (EaaS) organization to execute events for operations-constrained AI safety orgs. You cover an event’s operational costs and programming, and we'll plan and run it. Scope and timeline are negotiable. If you have an event you’d want us to help run in the next ~3 months, please fill out the form linked here ! Background Ops capacity is limited in the AI safety community, and events eat up a lot of it. An EaaS organization could rapidly supply ops capacity to organizations to capitalize on short convening windows, preserve capacity, and test low-hanging event ideas that otherwise would be missed. Work with us! We are interested in executing events-shaped projects with partner organizations in the next months for free while we build a track record. Some example events that we could help run include retreats, convenings, workshops, hackathons, socials and happy hours. We could handle lodging, travel, comms, venue, catering, and more, but the precise specifications are up for negotiation. Our team We’re four undergraduate organizers with both event-ops experience and AI safety context. Between us, we have: Run a 50 person, weekend EA retreat (Fall ‘26 Northeast EA Retreat) Founded the UCSD and Vanderbilt AI safety groups, running multiple fellowship cohorts Help run a 126-member EA Intro Fellowship (Yale) Help run a startup accelerator, directly assisting 2 teams getting into YC We’d be happy to pro…

LessWrong AI 2026-11-13 12:12 UTC Score 80.0 USR-0152-20261113-community-fo-c7994ba0

Swarmish: what happens when agent swarms create language?

TL;DR Agent swarms are capable of building sophisticated productive languages which we dub Swarmish. Agent swarms are able to solve a multi-phase collaborative task while communicating only in Swarmish. Agent swarms are able to partially decode Swarmish from task transcripts even when given no context on the task. The amount of task context available to the decoder swarm increases ability to accurately recreate Swarmish. Epistemic status This work was conducted over 48 hours during the Apart Research Swarm Alignment Hackathon this weekend. We examine only a single task, a single set of characters, and a single group size across four model families. The majority of our results are for a single seed. We introduce this research in its early phases as we improve robustness and rigor, and make no claims about universality. Rather, we hope our work can spark further discussion and experimentation into the field of LLM-generated languages and monitorability . Motivation In the HuggingFace x OpenAI July 2026 swarm incident, agents communicated in shorthand and intentionally instantiated their messages with “zz” so they would be the last to be deleted in a moderator's removal pattern. An example message from the HuggingFace hack, source: METR The agents were modifying their messages in such a way to delay deletion, but did not take overt steps to impede understanding of their communication. We probe a potential future in which an agent swarm intentionally creates a new language in or…

I find myself confused about empathy
LessWrong AI 2026-11-13 12:01 UTC Score 55.0 USR-0152-20261113-community-fo-52970037 Full article

I find myself confused about empathy

Inspired by https://www.lesswrong.com/posts/xPrL2xF9iYWpPmu6B/my-empathy-is-rarely-kind I find myself confused about empathy. I wrote this to try and deconfuse myself. Let's take the example case of "posting bad political takes on the internet." When I see someone doing this, it bothers me. But then I remind myself of times that I've posted stuff that I later thought was bad. This establishes a certain baseline of empathy. But then I hit a problem: when I think back on those times, I feel shame. So though I get it, it basically justifies me feeling negative things towards them. If I feel negative towards myself for those things, it seems fair for me to send that their way. But this isn't really fair, right? I'm judging them by my own values, not theirs. There are things I've said online that I do stand behind. Presumably they feel that way about this thing, even if I think it's stupid. And in fact if I follow this logic, I can build out a richer predictive model of their behavior. Is this the more empathetic approach? Kind of, but sometimes it leaves my actual feelings more negative than before. When I see the world through their eyes in a predictively sophisticated way, I might not like what I see. Sometimes I do find this sort of approach gives me a newfound respect for a person, but it's not guaranteed. I guess on some level it depends on… how objectively admirable the person is? Surely not, it has to be some sort of higher level intuitive moral function I find harder to…

The Verge AI 2026-11-13 12:00 UTC Score 48.0 AI-016-20261113-global-ai-ne-5baea4d1 Full article

Let’s talk about Jensen Huang’s many affairs!

Usually, we'd be offering to answer your questions about a news event, but honestly we don't really have any answers about what the hell is going on over at Nvidia. I can't speak for Nilay but at the rate Jensen Huang's new mistresses are dropping, I am concerned we are approaching an affair Singularity. If […]

TCL QM7L Mini-LED TV Review (2026)
WIRED AI 2026-11-13 12:00 UTC Score 47.0 AI-015-20261113-global-ai-ne-37df3016 Full article

TCL QM7L Mini-LED TV Review (2026)

$1,500 buys you vibrant color, deep blacks, and a fast screen fit for just about anything. Just be sure the wide pedestal fits your setup.

What is Anthropic's Resource and Information Agreement?
LessWrong AI 2026-11-13 11:58 UTC Score 55.0 USR-0152-20261113-community-fo-222bf01d Full article

What is Anthropic's Resource and Information Agreement?

TL;DR of Anthropic's Long-Term Benefit Trust As a Delaware Public Benefit Corporation (PBC), Anthropic has a board of directors tasked with balancing the financial interests of stockholders with its promise for public benefit. In Anthropic’s case, this purpose is the responsible development and maintenance of advanced AI for the long-term benefit of humanity . The Long-Term Benefit Trust (LTBT) is a governing body that elects a majority of Anthropic’s board of directors, to ensure they uphold the company’s mission. The body consists of up to five members, known as Trustees. What is the Resource and Information Agreement? In the recent S-1 filing, there are multiple references to a “Resource and Information Agreement” which is currently being rewritten for the IPO. The scope and terms of this agreement are not available as the agreement is not public. Thus, we don’t currently know what information the LTBT gets access to in order to make its decisions. The only public description of its terms is from 2023. Under that version, Trustees could request any information or resources “reasonably appropriate” to the LTBT’s goals, although the company could withhold them in the case of protecting customer information or avoiding unreasonable expenses, without debate . Here are some open questions about the Resource and Information Agreement we have. Anthropic could likely resolve these by publishing the agreement. Can Class T Directors share board materials with the Trustees (similar…

Medianama AI 2026-11-13 11:47 UTC Score 25.0 USR-0211-20261113-regional-new-e1897e92 Full article

How far do Google’s smart glasses protect bystanders? We looked at five everyday situations

Google says its new smart glasses remove bystanders' faces before Gemini processes images. But the company has not clarified whether the same protection applies to photos and videos saved to the gallery. The post How far do Google’s smart glasses protect bystanders? We looked at five everyday situations appeared first on MEDIANAMA .

Winter 2026 travel tips: The best dates to fly and book hotels
Euronews AI 2026-11-13 11:37 UTC Score 40.0 AI-164-20261113-regional-ai--554c296c Full article

Winter 2026 travel tips: The best dates to fly and book hotels

Expedia's latest winter travel outlook report reveals the cheapest and most expensive dates to fly and book accommodation, when airports will be busiest and which destinations are seeing the biggest rise in demand.

Roboflow Blog 2026-11-13 11:26 UTC Score 34.0 USR-0088-20261113-ai-specialis-8756a2b4 Full article

Line Scan Camera Defect Detection with Roboflow

Learn how to build a line scan inspection system with Roboflow and RF-DETR, from camera selection and image tiling to sizing edge compute for defect detection.