GoKawiil - Latest Tech News & Aggregated Headlines

Thinking Machines Lab wants to make AI models more consistent

techcrunch.com Maxwell Zeff 2025-09-30 01:30:11

There’s been great interest in what Mira Murati’s Thinking Machines Lab is building with its $2 billion in seed funding and the all-star team of former OpenAI researchers who have joined the lab. In a blog post published on Wednesday, Murati’s research lab gave the world its first look into one of its projects: creating AI models with reproducible responses. The research blog post, titled “Defeating Nondeterminism in LLM Inference,” tries to unpack the root cause of what introduces randomness i

Topics: ai lab machines research thinking

Shop Amazon

GLM 4.5 with Claude Code

news.ycombinator.com Unknown 2025-10-06 05:45:08

GLM Coding Plan — designed for Claude Code users, starting at $3/month to enjoy a premium coding experience! GLM-4.5 and GLM-4.5-Air are our latest flagship models, purpose-built as foundational models for agent-oriented applications. Both leverage a Mixture-of-Experts (MoE) architecture. GLM-4.5 has a total parameter count of 355B with 32B active parameters per forward pass, while GLM-4.5-Air adopts a more streamlined design with 106B total parameters and 12B active parameters. Both models sh

Topics: glm models performance reasoning thinking

Shop Amazon

OpenAI is testing "Thinking effort" for ChatGPT

bleepingcomputer.com Unknown 2025-10-17 13:26:42

OpenAI is working on a new feature called the Thinking effort picker for ChatGPT. With the Thinking effort picker, you can choose how hard ChatGPT can think. When a model thinks harder, it doesn't necessarily mean a great response. You might not want the model to think harder when you're not sure what to do tomorrow. However, you may want the model to think harder when you're working on a complex topic, such as econometrics, bond valuation, healthcare and so on. As spotted on X, OpenAI is te

Topics: 200 harder model openai thinking

Shop Amazon

People Are REALLY Mad at These AI Glasses That Record Everything Constantly

futurism.com Unknown 2025-10-25 00:13:24

Are you looking forward to a future of casual but supercharged surveillance, in which inconspicuous wearable devices record everything private you do — ostensibly in service of making you "super intelligent?" Evidently, readers, you are not. Users on social media have responded with horror and outrage to a pair of smart glasses developed by a startup called Halo that its creators, a pair of Harvard dropouts, claim will feed you live AI-powered insights while logging and transcribing every conv

Topics: ai glasses halo just thinking

Shop Amazon

Harvard Startup Says Its Smart Glasses Will Do "Vibe Thinking" for You

futurism.com Unknown 2025-10-28 18:00:17

We live in an age where the tech industry's core pitch is for you to outsource your brain's cognitive functions, be it to an AI chatbot or an app. The idea has now reached a new zenith. A startup called Halo is releasing a pair of smart glasses that will record and transcribe all your conversations and use it to beam you AI-powered insights. It'll remember details you forgot and recall what someone told you they like, the startup says, arming you with facts it looks up on the fly and answering

Topics: ai glasses make think thinking

Shop Amazon

Evaluating LLMs for my personal use case

news.ycombinator.com Graham King 2025-10-28 15:40:33

Most models are excellent, so cost and latency dominate. It’s great that AI can win maths Olympiads, but that’s not what I’m doing. I mostly ask basic Rust, Python, Linux and life questions. So I did my own evaluation. I gathered 130 real prompts from my bash history (I use command line tool llm). I had Qwen3 235B Thinking and Gemini 2.5 Pro group them into categories. They both chose very similar ones, broadly (with examples): Programming - “Write a bash script to ..” Sysadmin - “With curl

Topics: best deepseek models qwen3 thinking

Shop Amazon

GPT-5 usage limitations: what are they, how does this compare to GPT-4 family?

androidauthority.com Unknown 2025-10-31 01:00:51

Edgar Cervantes / Android Authority GPT-5 arrived a few weeks ago, though its rollout hasn’t been entirely smooth. While the model shows plenty of promise, its debut also meant the abrupt removal of every other GPT model from ChatGPT’s user-facing UI. Since then, some old models have returned, and there have been a few other changes to the way the system works. Furthermore, many of the initial GPT-5 usage limits have been temporarily enhanced since launch. Let’s dive in and take a closer look

Topics: gpt limits model models thinking

Shop Amazon

In the long run, LLMs make us dumber

news.ycombinator.com .Wp-Block-Post-Author Box-Sizing Border-Box Display Flex Flex-Wrap Wrap .Wp-Block-Post-Author__Byline Font-Size Margin-Bottom Margin-Top Width 2025-10-31 10:10:39

The comfort we get when offloading our cognitive load to LLMs is bad for us. Cognitive load should exist, and if we reduce it too much – if we stop thinking – we can actually unlearn how to think. Kids who always choose the easy route and copy their homework from other students eventually find themselves completely clueless about what’s going on in school. Someone who always lets their spouse handle all the bills and banking may one day be unable to manage even a simple payment on their own. A

Topics: cognitive group llm participants thinking

Shop Amazon

ChatGPT-5 Lets You Choose Your AI Model. These Are Your Options

cnet.com See Full Bio 2025-11-01 00:15:02

The biggest pushback after OpenAI announced its new GPT-5 model for ChatGPT came from devotees of older models who felt the new generative AI chatbot lacked the panache of its predecessors. Now you have more choices of pre-GPT-5 models (although you'll have to hunt for some of them) and better control over which components of GPT-5 handle your questions. OpenAI is still sorting through a somewhat rocky launch of GPT-5, led by complaints about the lack of model choices. The model has been antic

Topics: gpt model models openai thinking

Shop Amazon

5 reasons why GPT-5 is actually better than some of the older GPT models

androidauthority.com Unknown 2025-11-04 06:00:08

Joe Maring / Android Authority Recently, OpenAI has come under fire for GPT-5’s rocky launch. Many users have called it a step backward, citing a lack of personality and other tweaks that turned people off — sentiments echoed in our own GPT-5 review. Still, GPT-5 does improve on at least some of the previous legacy models. Before we dive in, it’s important to note that GPT-5 really does have less personality. It’s curt and to the point in nearly every interaction. This makes it much less usefu

Topics: 03 gpt models older thinking

Shop Amazon

OpenAI relaxes GPT-5 rate limit, promises to improve the personality

bleepingcomputer.com Unknown 2025-11-11 21:22:04

OpenAI is slowly addressing all concerns around GPT-5, including rate limits and now its personality, which has been criticized for being less affirmative. In a support document, OpenAI confirmed it has restored the older models for paid customers, so you can now use GPT4o, GPT o3, and more. You just need to use the model selector and choose one of the models under legacy models. In addition, GPT-5 automatically switches between Fast and Thinking, and you can also choose additional GPT-5 opti

Topics: gpt model offers openai thinking

Shop Amazon

OpenAI brings GPT-4o back as a default for all paying ChatGPT users, Altman promises ‘plenty of notice’ if it leaves again

venturebeat.com Carl Franzen 2025-11-17 09:11:16

Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now OpenAI is once again making GPT-4o — the large language model (LLM) that powered ChatGPT before last week’s launch of GPT-5 — a default option for all paying users, that is, those who subscribe to the ChatGPT Plus ($20 per month), Pro ($200 per month), Team ($30 per month), Enterprise, or Edu tiers, no longer requiring users to toggle on a

Topics: chatgpt gpt models thinking users

Shop Amazon

You can pick ChatGPT's older AI models again

engadget.com Unknown 2025-11-18 03:15:49

ChatGPT will now allow you to choose between several GPT-5 variants and previous OpenAI models. In a post on X, OpenAI CEO Sam Altman has revealed that the chatbot's model picker now comes with three GPT-5 choices: Auto, Fast and Thinking. He said most "will want Auto," which is most likely the standard version that's already a reasoning model. But users will be able to choose the fast-responding version or the Thinking version that implies it delivers longer, more comprehensive answers if they

Topics: 4o gpt openai thinking users

Shop Amazon

AI engineers reject Meta's $1.5 billion offers to build on their own terms

techspot.com Skye Jacobs 2025-12-06 20:03:00

Editor's take: As Big Tech and industry leaders dig ever deeper into their war chests to recruit AI talent, a profound question is emerging: Can money alone win the battle for the minds building tomorrow's most powerful technologies? Recent events suggest that, for a growing number of engineers and researchers, the answer is increasingly, and resoundingly, no. While not in the majority, many engineers are choosing to pass up unprecedented offers in favor of staying loyal to their mission, value

Topics: ai machines meta thinking work

Shop Amazon

AI engineers reject Meta's $1.5 billion offers to stay loyal to their mission

techspot.com Skye Jacobs 2025-12-07 08:03:00

Editor's take: As industry leaders dig ever deeper into their war chests to recruit AI talent, a profound question is emerging: Can money alone win the battle for the minds building tomorrow's most powerful technologies? Recent events suggest that, for a growing number of engineers and researchers, the answer is increasingly, and resoundingly, no. While not in the majority, many engineers are choosing to pass up unprecedented offers in favor of staying loyal to their mission, values, and the ch

Topics: ai machines meta thinking work

Shop Amazon

Ask HN: What are you working on? (July 2025)

news.ycombinator.com Unknown 2025-12-20 23:19:19

What are you working on? Do you have any new ideas you're thinking about?

Topics: ideas new thinking working

Shop Amazon

It’s Qwen’s summer: new open source Qwen3-235B-A22B-Thinking-2507 tops OpenAI, Gemini reasoning models on key benchmarks

venturebeat.com Carl Franzen 2025-12-22 11:47:28

Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now If the AI industry had an equivalent to the recording industry’s “song of the summer” — a hit that catches on in the warmer months here in the Northern Hemisphere and is heard playing everywhere — the clear honoree for that title would go to Alibaba’s Qwen Team. Over just the past week, the frontier model AI research division of the Chines

Topics: ai model models qwen3 thinking

Shop Amazon

I avoid using LLMs as a publisher and writer

news.ycombinator.com Tomáš Baránek 2026-01-03 06:51:59

Now for my more detailed arguments. Reason 1: I don’t want to become cognitively lazy In a recent study by MIT researchers (Your Brain on ChatGPT: Accumulation of Cognitive Debt When Using an AI Assistant for Essay Writing Task) demonstrated using LLMs when writing essays reduces the originality of the resulting work. More notably, when measured using an EEG, LLMs also diminish brain connectivity compared to when participants were allowed to use only their brains or a search engine. People who

Topics: llms text thinking using writing

Shop Amazon

Thinking Machines Lab Raises a Record $2 Billion, Announces Cofounders

wired.com Will Knight 2026-01-12 08:02:19

Thinking Machines Lab, an artificial intelligence company founded by top researchers who fled OpenAI, has raised a record $2 billion seed round that values the fledgling firm at $12 billion. The funding round was led by Andreessen Horowitz and included Nvidia, Accel, Cisco, and AMD—among others. The mammoth investment reflects the ultracompetitive race to build advanced AI systems, as well as the premium placed on top AI talent. It is the largest seed funding round in history. Thinking Machine

Topics: ai company machines openai thinking

Shop Amazon

Mira Murati’s Thinking Machines Lab is worth $12B in seed round

techcrunch.com Maxwell Zeff 2026-01-13 08:59:29

Thinking Machines Lab, the AI startup founded by OpenAI’s former chief technology officer Mira Murati, officially closed a $2 billion seed round led by Andreessen Horowitz on Monday, a company spokesperson told TechCrunch. The deal, which includes participation from NVIDIA, Accel, ServiceNow, CISCO, AMD, and Jane Street, values the startup at $12 billion, the spokesperson said. Several outlets reported in June that Thinking Machines Lab was close to closing this $2B funding round at a $10 bill

Topics: ai lab machines murati thinking

Shop Amazon

A new paradigm for AI: How ‘thinking as optimization’ leads to better general-purpose models

venturebeat.com Ben Dickson 2026-01-17 12:26:03

Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now Researchers at the University of Illinois Urbana-Champaign and the University of Virginia have developed a new model architecture that could lead to more robust AI systems with more powerful reasoning capabilities. Called an energy-based transformer (EBT), the architecture shows a natural ability to use inference-time scaling to solve comp

Topics: data ebts energy models thinking

Shop Amazon

Overclocking LLM Reasoning: Monitoring and Controlling LLM Thinking Path Lengths

news.ycombinator.com Roy Eisenstadt 2026-01-30 12:53:13

This work investigates how large reasoning models internally track their thinking progress and how such processes can be monitored and controlled. We focus on reasoning models that explicitly segment their computations using <think> and </think> tokens (e.g., DeepSeek-R1), allowing us to study the internal dynamics of the "thinking phase." 1. Monitoring the Thinking Phase We hypothesize that hidden states encode a token's relative position within the thinking phase. To test this, we collect hi

Topics: alpha reasoning theta thinking token

Shop Amazon

Ask HN: What Are You Working On? (June 2025)

news.ycombinator.com Unknown 2026-02-13 06:21:28

What are you working on? Any new ideas which you're thinking about?

Topics: ideas new thinking working

Shop Amazon

The Perils of 'Design Thinking'

news.ycombinator.com Celine Nguyen 2026-02-10 12:27:11

On the first day of a required class for freshman design majors at Carnegie Mellon, my professor stood in front of a lecture hall of earnest, nervous undergraduates and asked, “Who here thinks that design can change the world?” Several hands shot up, including mine. After a few seconds of silence, he advanced to the next slide of his presentation: a poster by the designer Frank Chimero that read, Design won’t save the world. Go volunteer at a soup kitchen, you pretentious fuck. My professor was

Topics: city design designers gram thinking

Shop Amazon

'Sticky thinking' hampers decisions in depression

news.ycombinator.com Unknown 2026-02-15 21:01:57

Though low mood is the best-known symptom of depression, indecisiveness and biased thinking are also common. But these problems, which can lead to delays in making important choices — and being more likely to make poor ones — receive less attention, write the authors of a recent paper in Emotion. Many of the theories that have been put forward to explain difficulties with decision-making are based on work suggesting that depressed people are not as responsive to rewards, or punishments, as thos

Topics: making participants sticky thinking vugt

Shop Amazon

What Apple's controversial research paper really tells us about LLMs

zdnet.com Sabrina Ortiz 2026-03-08 15:06:46

CHRISTOPH BURGSTEDT/SCIENCE PHOTO LIBRARY/Getty Generative AI models quickly proved they were capable of performing technical tasks well. Adding reasoning capabilities to the models unlocked unforeseen capabilities, enabling the models to think through more complex questions and produce better-quality, more accurate responses -- or so we thought. Last week, Apple released a research report called "The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the

Topics: model models paper reasoning thinking

Shop Amazon

I convinced HP's board to buy Palm and watched them kill it

news.ycombinator.com Phil Mckinney 2026-03-12 04:03:30

I've never shared this story publicly before—how I convinced HP's board to acquire Palm for $1.2 billion, then watched as they destroyed it while I was confined to bed recovering from surgery. This isn't just another tech failure analysis. I was the HP Chief Technology Officer who led the technical due diligence on Palm. I presented to Mark Hurd and the HP board, making the case for moving forward with the acquisition. I believed we were buying the future of mobile computing. Then I watched it

Topics: hp innovation technology thinking webos

Shop Amazon

I Convinced HP's Board to Buy Palm for $1.2B. I Watched Them Kill It in 49 Days

news.ycombinator.com Phil Mckinney 2026-03-12 19:03:30

I've never shared this story publicly before—how I convinced HP's board to acquire Palm for $1.2 billion, then watched as they destroyed it while I was confined to bed recovering from surgery. This isn't just another tech failure analysis. I was the HP Chief Technology Officer who led the technical due diligence on Palm. I presented to Mark Hurd and the HP board, making the case for moving forward with the acquisition. I believed we were buying the future of mobile computing. Then I watched it

Topics: hp innovation technology thinking webos

Shop Amazon

Latest Tech News

Thinking Machines Lab wants to make AI models more consistent

GLM 4.5 with Claude Code

OpenAI is testing "Thinking effort" for ChatGPT

People Are REALLY Mad at These AI Glasses That Record Everything Constantly

Harvard Startup Says Its Smart Glasses Will Do "Vibe Thinking" for You

Evaluating LLMs for my personal use case

GPT-5 usage limitations: what are they, how does this compare to GPT-4 family?

In the long run, LLMs make us dumber

ChatGPT-5 Lets You Choose Your AI Model. These Are Your Options

5 reasons why GPT-5 is actually better than some of the older GPT models

OpenAI relaxes GPT-5 rate limit, promises to improve the personality

OpenAI brings GPT-4o back as a default for all paying ChatGPT users, Altman promises ‘plenty of notice’ if it leaves again

You can pick ChatGPT's older AI models again

AI engineers reject Meta's $1.5 billion offers to build on their own terms

AI engineers reject Meta's $1.5 billion offers to stay loyal to their mission

Ask HN: What are you working on? (July 2025)

It’s Qwen’s summer: new open source Qwen3-235B-A22B-Thinking-2507 tops OpenAI, Gemini reasoning models on key benchmarks

I avoid using LLMs as a publisher and writer

Thinking Machines Lab Raises a Record $2 Billion, Announces Cofounders

Mira Murati’s Thinking Machines Lab is worth $12B in seed round

A new paradigm for AI: How ‘thinking as optimization’ leads to better general-purpose models

Overclocking LLM Reasoning: Monitoring and Controlling LLM Thinking Path Lengths

Ask HN: What Are You Working On? (June 2025)

The Perils of 'Design Thinking'

'Sticky thinking' hampers decisions in depression

What Apple's controversial research paper really tells us about LLMs

I convinced HP's board to buy Palm and watched them kill it

I Convinced HP's Board to Buy Palm for $1.2B. I Watched Them Kill It in 49 Days

About GoKawiil

Privacy

Advertising

Latest Tech News

Trending Topics

Hot Now

Popular

Emerging

About GoKawiil

Privacy

Advertising