Nathan Lambert

Featured in: Favicon

medium.com Favicon

substack.com Favicon

jamanetwork.com Favicon

towardsdatascience.com Favicon

robohub.org Favicon

exponentialview.co

Articles

Llama 4 is a technical triumph and a strategic stumble

2 weeks ago | ignorance.ai | Charlie Guo |Nathan Lambert

Meta's Llama 4 launch should have been a triumphant moment for open-weight AI. Instead, it's become a showcase of how technical excellence can be undermined by rushed execution and communication missteps (and why taking shortcuts may be the natural response to increasing competitive pressures). In the days since Meta (unexpectedly) released its newest AI model family over the weekend, the company's narrative has been at odds with the community's reception.
ðŸ“ˆ Data to start your week

1 month ago | exponentialview.co | Azeem Azhar |Nathan Lambert |Joachim Klement

Hi,Hereâ€™s your Monday round-up of data driving conversations this week â€” all in less than 250 words. A Chinese discount. Baiduâ€™s new multimodal foundation model, ERNIE 4.5, outperforms OpenAIâ€™s GPT-4.5 on several benchmarksâ€”while being 99% cheaper. An open alt. The Allen Institute released OLMo 2 32B, the first fully open-source1 GPT-4-class model. (via )Solar capacity.
Export Controls and HBM

Dec 10, 2024 | chinatalk.media | Jordan Schneider |Nathan Lambert

ChinaTalk is hiring for a dedicated China AI lab analyst. Chinese fluency and a technical background are strongly preferred. Apply here!We’ve got a new show up on the podcast feed with of the Interconnects Substack talking through the biggest AI stories of this year and next. Listen in on Apple Podcasts or Spotify. Today we’re running a guest piece from Ray Wang, a Washington-based analyst.
Building on evaluation quicksand

Oct 16, 2024 | interconnects.ai | Nathan Lambert

Other than the pace of progress on individual evaluations being extremely high, the landscape of evaluating leading language models has not changed substantially in the last year. The biggest change is the required level of detail in reporting results, where more information must be communicated to contextualize the capabilities of your models. The new types of models that OpenAI's o1 heralds welcome a new axis to this, evaluation-time compute.
A post-training approach to AI regulation with Model Specs

Sep 9, 2024 | interconnects.ai | Nathan Lambert

The central point of AI regulation and policy over the last few years, everything from the Biden Executive Order to California’s SB 1047 bill, has been model size. The most common tool for AI enforcement proportional to model size has been thresholds that kick in once an AI system uses more than a certain amount of compute (or money) to be trained. The use of thresholds for regulation is the subject of substantial pushback and is likely fading in relevance due to it.
On Nous Hermes 3 and classifying a "frontier model"

Aug 16, 2024 | interconnects.ai | Nathan Lambert

Edit 1, 16 August: I made a few errors in this original post, which makes the Nous Research team, and by extension the EluetherAI evaluation harness, look bad. This isn’t my intention. As a summary:The Nous report is clear on their evaluation recipe. The open question of how best to evaluate Llama 3.1 models still holds. I’ve changed the title of the post. I’ve toned down the text in a few places to better reflect this. Sorry for the error.
Why OpenAI's CriticGPT Might be Significant

Jul 2, 2024 | ai-supremacy.com | Michael Spencer |Nathan Lambert |Cameron Wolfe

Hey Everyone, One of the major problems with LLMs is their hallucinations and tendency not be human aligned properly. On June 27th, OpenAI announced what it calls CriticGPT. While Anthropic is six years younger than OpenAI (2021 vs. 2015 respectively) now in July, 2024, Claude 3.5 is known to be better human aligned than ChatGPT and better apple to follow human instructions. Clearly OpenAI has some work to do.
AI for the rest of us

Jun 14, 2024 | interconnects.ai | Nathan Lambert

By being quiet in the race for the biggest and most rad foundation models, many people assumed that Apple didn’t have anything to contribute to the AI race. After their new announcements, many in the heart of the AI labs still will argue this, forecasting that . Apple is betting that AI follows paths paved by previous technological revolutions: incremental progress to transformational results.
We aren’t running out of training data, we are running out of open training data

May 29, 2024 | interconnects.ai | Nathan Lambert

For months we’ve been getting stories about how the leading teams training language models (LMs) are running out of data for their next generation of models — vaguely insinuating a struggle for big tech’s darling industry with no strategic claims beyond the fact that the second derivative on training dataset size is negative.
Cold War Lessons for AI Dialogue

May 22, 2024 | chinatalk.media | Jordan Schneider |Nathan Lambert |Lawrence Freedman

I posted a short show with discussing the technological and political implications of the GPT4o and Google’s dev day. Check it out in your favorite podcast app. Last week, a Biden Administration delegation flew to Geneva to talk AI with representatives from Beijing. Many AI commentators have analogized to US-USSR nuclear negotiations, but few seems to have really read all that much history. What can we learn from actual Cold War nuclear history relevant to today’s AI discussions?