AXRP (pronounced axe-urp) is the AI X-risk Research Podcast where I, Daniel Filan, have conversations with researchers about their papers. We discuss the paper, and hopefully get a sense of why it's been written and how it might reduce the risk of AI causing an existential catastrophe: that is, permanently and drastically curtailing humanity's future potential. You can visit the website and read transcripts at axrp.net.
…
continue reading
Every human being deserves to live their life to the fullest. Yet, every single day, drug addiction robs people and families of that opportunity. That problem is the driving force of this podcast. We provide a helping hand for people going through recovery; and a credible source for the truth about beating addiction. Every episode is a judgment-free space for healing, hope, and empowerment.
…
continue reading
Your new favorite sports show. Tune in with Danache and special guests as she breaks down NFL hot topics, dissecting everything NFL on and off the field, game analysis, fantasy tips, game picks and more.
…
continue reading
![Artwork](/static/images/128pixel.png)
1
35 - Peter Hase on LLM Beliefs and Easy-to-Hard Generalization
2:17:24
2:17:24
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:17:24
How do we figure out what large language models believe? In fact, do they even have beliefs? Do those beliefs have locations, and if so, can we edit those locations to change the beliefs? Also, how are we going to get AI to perform tasks so hard that we can't figure out if they succeeded at them? In this episode, I chat to Peter Hase about his rese…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
34 - AI Evaluations with Beth Barnes
2:14:02
2:14:02
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:14:02
How can we figure out if AIs are capable enough to pose a threat to humans? When should we make a big effort to mitigate risks of catastrophic AI misbehaviour? In this episode, I chat with Beth Barnes, founder of and head of research at METR, about these questions and more. Patreon: patreon.com/axrpodcast Ko-fi: ko-fi.com/axrpodcast The transcript:…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
33 - RLHF Problems with Scott Emmons
1:41:24
1:41:24
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:41:24
Reinforcement Learning from Human Feedback, or RLHF, is one of the main ways that makers of large language models make them 'aligned'. But people have long noted that there are difficulties with this approach when the models are smarter than the humans providing feedback. In this episode, I talk with Scott Emmons about his work categorizing the pro…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
32 - Understanding Agency with Jan Kulveit
2:22:29
2:22:29
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:22:29
What's the difference between a large language model and the human brain? And what's wrong with our theories of agency? In this episode, I chat about these questions with Jan Kulveit, who leads the Alignment of Complex Systems research group. Patreon: patreon.com/axrpodcast Ko-fi: ko-fi.com/axrpodcast The transcript: axrp.net/episode/2024/05/30/epi…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
31 - Singular Learning Theory with Daniel Murfet
2:32:07
2:32:07
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:32:07
What's going on with deep learning? What sorts of models get learned, and what are the learning dynamics? Singular learning theory is a theory of Bayesian statistics broad enough in scope to encompass deep neural networks that may help answer these questions. In this episode, I speak with Daniel Murfet about this research program and what it tells …
…
continue reading
![Artwork](/static/images/128pixel.png)
1
30 - AI Security with Jeffrey Ladish
2:15:44
2:15:44
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:15:44
Top labs use various forms of "safety training" on models before their release to make sure they don't do nasty stuff - but how robust is that? How can we ensure that the weights of powerful AIs don't get leaked or stolen? And what can AI even do these days? In this episode, I speak with Jeffrey Ladish about security and AI. Patreon: patreon.com/ax…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
29 - Science of Deep Learning with Vikrant Varma
2:13:46
2:13:46
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:13:46
In 2022, it was announced that a fairly simple method can be used to extract the true beliefs of a language model on any given topic, without having to actually understand the topic at hand. Earlier, in 2021, it was announced that neural networks sometimes 'grok': that is, when training them on certain tasks, they initially memorize their training …
…
continue reading
![Artwork](/static/images/128pixel.png)
1
28 - Suing Labs for AI Risk with Gabriel Weil
1:57:30
1:57:30
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:57:30
How should the law govern AI? Those concerned about existential risks often push either for bans or for regulations meant to ensure that AI is developed safely - but another approach is possible. In this episode, Gabriel Weil talks about his proposal to modify tort law to enable people to sue AI companies for disasters that are "nearly catastrophic…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
27 - AI Control with Buck Shlegeris and Ryan Greenblatt
2:56:05
2:56:05
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:56:05
A lot of work to prevent AI existential risk takes the form of ensuring that AIs don't want to cause harm or take over the world---or in other words, ensuring that they're aligned. In this episode, I talk with Buck Shlegeris and Ryan Greenblatt about a different approach, called "AI control": ensuring that AI systems couldn't take over the world, e…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
26 - AI Governance with Elizabeth Seger
1:57:13
1:57:13
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:57:13
The events of this year have highlighted important questions about the governance of artificial intelligence. For instance, what does it mean to democratize AI? And how should we balance benefits and dangers of open-sourcing powerful AI systems such as large language models? In this episode, I speak with Elizabeth Seger about her research on these …
…
continue reading
![Artwork](/static/images/128pixel.png)
1
25 - Cooperative AI with Caspar Oesterheld
3:02:09
3:02:09
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
3:02:09
Imagine a world where there are many powerful AI systems, working at cross purposes. You could suppose that different governments use AIs to manage their militaries, or simply that many powerful AIs have their own wills. At any rate, it seems valuable for them to be able to cooperatively work together and minimize pointless conflict. How do we ensu…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
24 - Superalignment with Jan Leike
2:08:29
2:08:29
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:08:29
Recently, OpenAI made a splash by announcing a new "Superalignment" team. Lead by Jan Leike and Ilya Sutskever, the team would consist of top researchers, attempting to solve alignment for superintelligent AIs in four years by figuring out how to build a trustworthy human-level AI alignment researcher, and then using it to solve the rest of the pro…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
23 - Mechanistic Anomaly Detection with Mark Xu
2:05:52
2:05:52
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:05:52
Is there some way we can detect bad behaviour in our AI system without having to know exactly what it looks like? In this episode, I speak with Mark Xu about mechanistic anomaly detection: a research direction based on the idea of detecting strange things happening in neural networks, in the hope that that will alert us of potential treacherous tur…
…
continue reading
Very brief survey: bit.ly/axrpsurvey2023 Store is closing in a week! Link: store.axrp.net/ Patreon: patreon.com/axrpodcast Ko-fi: ko-fi.com/axrpodcast
…
continue reading
![Artwork](/static/images/128pixel.png)
1
22 - Shard Theory with Quintin Pope
3:28:21
3:28:21
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
3:28:21
What can we learn about advanced deep learning systems by understanding how humans learn and form values over their lifetimes? Will superhuman AI look like ruthless coherent utility optimization, or more like a mishmash of contextually activated desires? This episode's guest, Quintin Pope, has been thinking about these questions as a leading resear…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
21 - Interpretability for Engineers with Stephen Casper
1:56:02
1:56:02
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:56:02
Lots of people in the field of machine learning study 'interpretability', developing tools that they say give us useful information about neural networks. But how do we know if meaningful progress is actually being made? What should we want out of these tools? In this episode, I speak to Stephen Casper about these questions, as well as about a benc…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
20 - 'Reform' AI Alignment with Scott Aaronson
2:27:35
2:27:35
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
2:27:35
How should we scientifically think about the impact of AI on human civilization, and whether or not it will doom us all? In this episode, I speak with Scott Aaronson about his views on how to make progress in AI alignment, as well as his work on watermarking the output of language models, and how he moved from a background in quantum complexity the…
…
continue reading
Danache and Jerry rank their favorite off season moves so far. The Lamar Jackson Saga continues, AFC North needs and all the latest around the league.Danache & Jeremy
…
continue reading
URP 2023 03 22Danache & Jeremy
…
continue reading
Are you ready for some football? With the Jets on the market for a quarterback, it looks like Lamar Jackson's ball game! Dan Snyder, the Colts, the Titans, the Texans, the Bucks, and even the Patriots all need a quarterback. Who will be the one to take the lead? Tune in to find out!Danache & Jeremy
…
continue reading
03-15-23Danache & Jeremy
…
continue reading
![Artwork](/static/images/128pixel.png)
1
MONEY, MONEY, MONEY w. Special Guest Lawrence "Bam Bam"
1:04:06
1:04:06
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:04:06
Tune is as the team discuss NFL combine, franchise tags, Derek Carr, Lamar Jackson, Aaron Rodgers, and break down what each NFL team needs, starting with the AFC East.Danache & Jeremy
…
continue reading
This week Danache and Jerry talk about the lack of black coaches in the NFL, the NFL's unwillingness to hire Eric Bieniemy as a head coach after years of showing he's a capable OC, the Giants look like their heading in a mess of their own. Daniel Jones? Saquon? or both? and more..tune in.Danache & Jeremy
…
continue reading
Danache and Jerry dive into Super Bowl LVII, the history and importance of the black quarterback, what a win or loss could mean for either Patrick Mahomes and Jalen Hurts. Are these two on the path to becoming one of the greatest to ever play the game? Andy Reid's legacy as a coach. Updates on some coaching moves around the league, LamarJackson and…
…
continue reading
Store: https://store.axrp.net/ Patreon: https://www.patreon.com/axrpodcast Ko-fi: https://ko-fi.com/axrpodcast Video: https://www.youtube.com/watch?v=kmPFjpEibu0
…
continue reading
![Artwork](/static/images/128pixel.png)
1
19 - Mechanistic Interpretability with Neel Nanda
3:52:47
3:52:47
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
3:52:47
How good are we at understanding the internal computation of advanced machine learning models, and do we have a hope at getting better? In this episode, Neel Nanda talks about the sub-field of mechanistic interpretability research, as well as papers he's contributed to that explore the basics of transformer circuits, induction heads, and grokking. …
…
continue reading
Danache is joined by the Jospeh brothers and they break down what happened in the division championship games, the 49ers and their QB nightmares, coaching decisions that just make sense, Super Bowl preview and much more. And of course a heartfelt farewell to BradyDanache & Jeremy
…
continue reading
Danache and Jerry are joined by What Just Happened Sports founder Melissa Anthony. Tune in as they discuss, her journey, the importance of equipping and empowering women with sports knowledge, preview of this weekends conference championship games, and answering the question of what’s wrong with the Buffalo Bills?…
…
continue reading
Danache and Jerry talk super wild card weekend, Aaron Rodgers, does he stay, does he go? Lovie Smith and those good ol' Texans. Houston has a problem and it wasn't Lovie Smith. Game picks and more.Danache & Jeremy
…
continue reading
Danache is back with special guests as the team discusses Buffalo Bills Safety Damar Hamlin, his road to recovery, the NFL community's next steps and more.Danache & Jeremy
…
continue reading
Only thing better than 2 is 3. Jon, Jerry, & Ed talk Deion Sanders and the weight of the move. Are the Cowboys an OBJ away from best team in the league? The Bucs are all but guaranteed to make the playoffs but what's the point? The Bengals could be the most slept on team in the league and the 49ers might just be alright with Mr Irrelevant…
…
continue reading
The clock is ticking in the NFL & strength of schedule MATTERS. Jonathan & Jerry talk about the great differentiator between contenders and pretenders. You're QB can be YOUR guy, but not THE guy. The decisions that just have to be made, and welcome the fresh faces of the Coach of the Year race.Danache & Jeremy
…
continue reading
Thanksgiving is here and we preview the best Thanksgiving line up we may have possibly seen in years. Zach Wilson benched..what are the JETS going to do? We might have an idea. Tune in.Danache & Jeremy
…
continue reading
The NFL season is starting to turn up and the next few weeks can change everything. The AFC and NFC East have turned to probably the best divisons in football. Kirk Cousins isn't looking like a joke anymore (well 1 pm kirk that is). Tune is as we cover whats been going on around the league.Danache & Jeremy
…
continue reading
Jon & Ed break down why Jeff Saturday makes sense and why it absolutely doesn't. Why the AFC East is the best division in the league. The biggest disappointments of the league and predictions for Week 10!Danache & Jeremy
…
continue reading
The Jets & Giants are back. Did they get to t he party. a little too early or are they right on time and here to stay?Danache & Jeremy
…
continue reading
![Artwork](/static/images/128pixel.png)
1
All My Children
1:00:20
1:00:20
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:00:20
Jon & Ed “Mr ESPN” break down Week 6, why you might NOT want to pay Lamar if you’re the Ravens, and how a lot of these QB’s share a striking resemblance to a certain Super Bowl winner’s gameplayDanache & Jeremy
…
continue reading
I have a new podcast, where I interview whoever I want about whatever I want. It's called "The Filan Cabinet", and you can find it wherever you listen to podcasts. The first three episodes are about pandemic preparedness, God, and cryptocurrency. For more details, check out the podcast website (thefilancabinet.com), or search "The Filan Cabinet" in…
…
continue reading
Jeremy is back!!!! These two jump right in and discuss all things week 4 and preview what’s to come for week 5. #packers #nfc east #ny giants #green bay packers #all the smoke #joy Taylor #taylor rooks #skip bayless #london #new England patriots #bill belichick #josh McDaniels #miami #miami Dolphins #tua #teddy bridgewater #detriot #detriot lions #…
…
continue reading
Special Guest @UGOBOY88 joins the. show and we talk about it all. Is there a new king in the AFC South? The AFC West clearly isn't living up to expectations (yet) and are the Cowboys back? This and sooooo much more #dallas cowboys #dak Prescott # Tom Brady #the Pivot #under Review #under review podcast #undisputed #first Take #jaguars #miami #dolph…
…
continue reading
Tune in as Danache and special guest Ed discuss discuss everything from the AFC juggernauts, to Kirk Cousins dismal performance. Is it time for the Vikings to move on? One person seems to know so.Danache & Jeremy
…
continue reading
Danache is back and she is joined b y special guest, Ravens Insider Ms. Gina from SheTalksFootball. These two women talk week 1, Lamar Jackson and that money, AFC West and who really the Kings of that division, Jimmy G and more. Of course fantasy tips and weekly predictions.Danache & Jeremy
…
continue reading
![Artwork](/static/images/128pixel.png)
1
18 - Concept Extrapolation with Stuart Armstrong
1:46:19
1:46:19
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:46:19
Concept extrapolation is the idea of taking concepts an AI has about the world - say, "mass" or "does this picture contain a hot dog" - and extending them sensibly to situations where things are different - like learning that the world works via special relativity, or seeing a picture of a novel sausage-bread combination. For a while, Stuart Armstr…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
17 - Training for Very High Reliability with Daniel Ziegler
1:00:59
1:00:59
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:00:59
Sometimes, people talk about making AI systems safe by taking examples where they fail and training them to do well on those. But how can we actually do this well, especially when we can't use a computer program to say what a 'failure' is? In this episode, I speak with Daniel Ziegler about his research group's efforts to try doing this with present…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
16 - Preparing for Debate AI with Geoffrey Irving
1:04:49
1:04:49
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:04:49
Many people in the AI alignment space have heard of AI safety via debate - check out AXRP episode 6 (axrp.net/episode/2021/04/08/episode-6-debate-beth-barnes.html) if you need a primer. But how do we get language models to the stage where they can usefully implement debate? In this episode, I talk to Geoffrey Irving about the role of language model…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
15 - Natural Abstractions with John Wentworth
1:36:30
1:36:30
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:36:30
Why does anybody care about natural abstractions? Do they somehow relate to math, or value learning? How do E. coli bacteria find sources of sugar? All these questions and more will be answered in this interview with John Wentworth, where we talk about his research plan of understanding agency via natural abstractions. Topics we discuss, and timest…
…
continue reading
Danache is joined by Jonathan and they dive into the NBA playoffs. Grizzlies/Warriors a new found rivalry?!?! Best team in the AFC West, best team in the AFC North, what players are we looking forward to seeing this upcoming season, and what is up with these notable free agents? Julio Jones, Odell Beckham, Jarvis Landry, and more.…
…
continue reading
Why no one wants Baker..right now anyway. Cardinals pick up Kyler Murray’s 5th year option. AFC is on fire and the NFC looks like prime real estate, who is ready to take it? Brian Flores lawsuit update. Deshaun Watson, the Browns, and what needs to happen for them to move on. And of course the NFL draft. Twitter: @U_goboy @mzdanache IG: @ugoboy88 @…
…
continue reading
Did you miss us?! We’re back for Season 3 and we’re kicking this season off with our guy Jon hosting Eddy Mr ESPN, where we chop it up and touch on all things sports from the new Rooney rule, what the MLS has to do to be successful, black ownership in sports, the state of the NBA playoffs, and much more. Don’t wanna miss it!! Ed/eddyjoseph_pes eddy…
…
continue reading
![Artwork](/static/images/128pixel.png)
1
14 - Infra-Bayesian Physicalism with Vanessa Kosoy
1:47:31
1:47:31
Відтворити пізніше
Відтворити пізніше
Списки
Подобається
Подобається
1:47:31
Late last year, Vanessa Kosoy and Alexander Appel published some research under the heading of "Infra-Bayesian physicalism". But wait - what was infra-Bayesianism again? Why should we care? And what does any of this have to do with physicalism? In this episode, I talk with Vanessa Kosoy about these questions, and get a technical overview of how inf…
…
continue reading