Exploring new patents from OpenAI that seemingly no other outlet else spotted, plus an exclusive interview with Gavin Uberti, CEO of Etched AI, on how verification and this kind of mass sampling could push models to the state of the art in intelligence.
Gavin Uberti - Etched.AI (they are hiring)
OpenAI Patent 1: https://www.freepatentsonline.com/y2024/0020116.html
Patent 2:
2024-02-02 13:48:04 +0000 UTC
View Post
- Thoughts on a new paper showing GPT-4 Turbo acing my favourite Reading Comprehension challenge
- Are humans just ‘superior pattern processors’, and would that reframe your view of AGI?
- On Optimism Bias, aka ‘the cameraman never dies’
- What is the 'Reading Comprehension' moment for you?
- Bullish on Humans Having an Economic Role indefinitely
- Inviting Chris Miller of Chip Wars
GPT-4 Turbo Reading Comprehension: 2024-01-31 10:22:36 +0000 UTC
View Post
Lots of articles recently in the New York Times, Washington Post, Forbes and MIT about whether LLMs boost productivity for white-collar workers. Well, I've dug into the details behind 5 of the papers frequently cited, interviewing co-authors [Edward McFowland III from Harvard Business School] and revealing what no other outlet did. Plus, test yourself on a task that GPT-4 actually worsened productivity for, learn about the Simpsons Paradox, and hear what the IMF thinks about the future ...
2024-01-18 16:37:21 +0000 UTC
View Post
London-based Cybersecurity Consultant and fellow AI Insider (and YouTuber) Donato Capitella, gives us unique insight into what he works on professionally to avoid - prompt injections for large language models. Whether you haven't heard of a Prompt Injection, or work on avoiding such attacks daily, this video will have something to offer!
Slides: https://llm-chronicles.com/prompt-injection-talk/
2024-01-18 16:09:21 +0000 UTC
View Post
Drawing on 3 exclusive interviews and weeks of research, here is my uncovering of 4 underappreciated trends as we debate whether embodiment is needed for AGI.
3 Exclusive Interviews +
Sanctuary AI: https://time.com/collection/best-inventions-2023/6326928/sanctuary-ai-phoenix/
Sanctuary AI CEO:
2024-01-04 08:00:04 +0000 UTC
View Post
How I work, what I read, starting the channel, goals and more ...
OG AI Explained: https://www.youtube.com/watch?v=sujU5o3Bfys
Arxiv: https://arxiv.org/
Hypothesis Chrome Extension: https://chr...
2024-01-03 15:05:38 +0000 UTC
View Post
2024 will be a big year both for AI and for AI Insiders. Will be doing a New Year video on the main channel, outlining why I think AI advances are still being underestimated, and getting '24 off to a galloping start on here with a set of videos exclusively for AI Insiders in the first two weeks of January, including: a video on below-the-radar trends ft. Nvidia stars Jim Fan & Guanzhi Wang and Google DeepMind's Pannag Sanketi, a survey of Productivity Boosts from AI ft. Harvard Business S...
2023-12-31 17:54:30 +0000 UTC
View Post
An exclusive interview with Microsoft's Sébastien Bubeck, co-author of Sparks of AGI and the Phi series of models. From practical tips to AGI timelines, AI on a phone to whether models can reason, it was a great chat and it even ended on a powerful note regarding the future of scaling language models.
Exclusive Interview +
Bubeck: https://en.m.wikipedia.org/wiki/S%C3%A9bastien_Bubeck
2023-12-17 19:42:06 +0000 UTC
View Post
Speaking to Sebastien Bubeck in an hour, for AI Insiders. Bubeck is Sr. Principal Research Manager at Microsoft, co-author of Sparks of AGI and Phi-2 (Textbooks are All You Need) and Yann LeCun debater: A New Kind of Intelligence. A bit late notice I know, timing was just confirmed, but wanted to give Insiders a chance to suggest questions. No guarantees, but will strongly consider those upvoted! V...
2023-12-15 17:24:22 +0000 UTC
View Post
The debut video for the Insiders Arena, where guest experts - hopefully including you! - will bring in insight on concepts in AI, trends, or simply a cool coding project. This video is from none other than Swyx, made for AI Insiders exclusively.
Swyx is a developer/founder/angel investor, GitHub Star, Stripe Community Expert and helped run the React subreddit for over 200,000 developers.
He'll explain why we need the new term ‘AI Engineer’, why it’s smart to be active on ...
2023-12-12 13:04:57 +0000 UTC
View Post
As you guys may have noticed, I am now making Patreon a major focus of my career. That means more posts on Legendary Supporters, more regular updates and the whole new tier AI Insiders.
What is AI Insiders? Aside from being my way of keeping the channel sustainable and thriving into 2024, it's my outlet for creative content beyond the main AI Explained channel. That means a new podcast, exclusive AI Explained-style videos (one released tonight called AGI timelines includes snippets fro...
2023-12-08 23:34:56 +0000 UTC
View Post
Drawing on interviews with half a dozen top experts at companies like Nvidia, Google DeepMind and Microsoft, as well as a host of other experts in the field, I'll give you their - and my - timelines for artificial general intelligence. By the end of the video, I want you to have a much better sense of what scale is needed, and what the approximate consensus is.
Links:
6 Exclusive Interviews +
Metaculus Predictions: 2023-12-08 21:08:46 +0000 UTC
View Post
Possibly the most important question for the future of AI - when will models reason from first principles and break down the final barrier to AGI? Drawing on 3 (!) expert interviews including Professor Rao, a thought leader on this question, to present you with the key debate at the heart of the fate of LLMs.
Slides: https://docs.google.com/presentat...
2023-12-05 18:30:27 +0000 UTC
View Post
Concrete methods to reduce hallucinations, and an exploration of the hidden biases of models that I am 99.8% confident you will learn something new from, but a deep dive into RepE, a stunning new method from Andy Zou, who, surprise-surprise, I managed to interview for this video! Great for practitioners and anyone looking to understand more about the secrets of LLMs.
Slides:
2023-12-05 18:00:12 +0000 UTC
View Post
Featuring expert lead authors from Google DeepMind, Microsoft and more, we'll explore what the near term future of prompting will look like. We'll discuss how we might soon be getting Auto-SmartGPT and prompt's made by LLM's and optimised even beyond human recognition. With the authors of AutoGen we'll touch on the importance of System Prompts and last but not least discuss with a New York Times featured ML researcher how even Inflection Pi can be jailbroken with the right prompts.
2023-12-05 17:22:53 +0000 UTC
View Post
Whether you're new to LLMs or consider yourself a Prompt Engineer, here is a 30 min tour of my most-used Prompt Hacks, plus expert guest insight from Google DeepMind's Tim Rocktäschel. Covering everything from prompt chaining to self consistency, rephrase and respond to EmotionPrompt.
Slides: https://docs.google.com/presentation/d/1IlOXhv4JUX7SOmPaMK...
2023-12-05 16:54:43 +0000 UTC
View Post
Not like my normal videos! This is a 40 min philosophical stream of consciousness on some of the ramifications of actually achieving AGI. Everyone talks about getting there, very few talk about what happens when we are there. I start with thoughts on a post-labour world, the three centers of gravity (traditionalists, explorers and enjoyers), CERN as a scout and much more. Enjoy on a walk, a run, a drive or while high ... :)
2023-12-05 15:08:27 +0000 UTC
View Post
The two songs featured in my latest video, in full, and only on Patreon! Got more goodies coming for this place.
2023-10-23 14:05:54 +0000 UTC
View Post
It is strange being a full-time YouTuber, each day is unpredictable. One Tuesday you are chatting with a high-up working on the final stages of the EU AI Act, then Wednesday you are examining a weird LLM quirk [try copying 100 of the letter 'a': e.g. 'a a a a a a' on GPT 3.5 - it's super weird but not sufficiently so for a full video]. One day being sent free advanced copies of AI books in the mail [quite a good one actually], then next day doing a collab with a multi-million sub Youtub...
2023-08-03 13:44:38 +0000 UTC
View Post
Thank you all once again for your support.
First update. I have turned down probably 100 sponsorship offers in the last month, including one for up to six figures, spread over a few months, due to not fully believing in the product. Old fashioned, I know, but I have to believe in something to say it, or even promote it. I take the trust of my audience quite seriously and though I will make mistakes (and have done) they should always be honest ones. It is due to your support that I have ...
2023-05-06 12:48:56 +0000 UTC
View Post
Just wanted to thank every Patron who supported me early on. It has really motivated me to keep making videos. I look forward to repaying you all somehow.
:)
Philip (definitely not GPT 5 from the future)
2023-04-02 20:07:44 +0000 UTC
View Post
