Podcast terbaik Bluedot Impact (2024)

1
S3 Ep60: Courses on AI Safety and Biosecurity with Will Saunter, Co-Founder and Biosecurity Lead at BlueDot Impact 30:01

23d ago30:01

30:01

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Will Saunter, Co-Founder and Biosecurity Lead at BlueDot Impact. The social enterprise offers 12-week, part-time courses in AI Alignment, AI Governance and Biosecurity. Alumni have gone on to pursue high impact ventures in AI and biosecurity, such…

1
S3 Ep63: Personalised AI Tools with Cien Solon, co-founder and CEO of LaunchLemonade and Director of Scale That Thing! 31:01

2d ago31:01

31:01

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Cien Solon, co-founder and CEO of LaunchLemonade and Director of Scale That Thing!. AI advocate Cien Solon is fighting to make AI a tool at everyone’s disposal - an “AI for All” as Cien so pithily calls it. To help individuals and organisations ha…

1
S3 Ep62: How AI can Better Represent Voters in Parliament with Steve Endacott, Chairman of Neural Voice and Neural River 33:15

9d ago33:15

33:15

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Steve Endacott, Chairman of Neural Voice and Neural River. Neural Voice is a spin-off from incubator Neural River, a genAI platform enabling businesses to develop customer-assistance voicebots. Tune into this episode now to hear Steve how Steve le…

1
S3 Ep61: AI in Bioacoustics with Mickey Pardo, research fellow at Cornell University 27:38

16d ago27:38

27:38

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Mickey Pardo, research fellow at Cornell University and former Postdoctoral Associate at Colorado State University. Mickey studied the rumbles of savannah elephants in partnership with Save the Elephants, discovering that elephants’ naming system …

1
S3 Ep59: Transforming Data for Educators with AI with Ben Dodson, Co-Founder & CEO of Doowii 44:51

1M ago44:51

44:51

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Ben Dodson, co-founder and CEO of Doowii. Helping teachers harness AI to better understand and address their students’ needs, Doowoi helps non-technical users access data they previously couldn't, improving workflows and decision-making. Tune into…

1
S3 Ep58: Reshaping Biosecurity with Kevin Flyangolts, Founder & CEO of Aclid 30:10

1M ago30:10

30:10

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Kevin Flyangolts, Founder & CEO of Aclid. With biotechnology becoming increasingly digital and automated, biosecurity guidelines are constantly changing. Aclid enables end-to-end compliance for gene synthesis providers by automating biosecurity sc…

1
S3 Ep57: Revolutionising Machine Learning for Research & Development with Brad Magnetta, Co-Founder of modlee 34:10

1M ago34:10

34:10

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Brad Magnetta, co-founder of modlee. With a growing demand for in-house AI and ML solutions, modlee automates, simplifies and pools insights about machine learning development. Tune into this episode now to hear Brad discuss the distinction betwee…

1
S3 Ep56: Exploring Sextech for Good with Sophie Rossi, Founder & CEO of HedonX 28:42

2M ago28:42

28:42

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Sophie Rossi, founder and CEO of HedonX. With the femtech and sextech industries seeing a surge in sales during the pandemic, HedonX’s soon-to-launch range of intimate wellness products will satisfy this demand all while promoting the use of ethic…

1
S3 Ep55: Building Wealth for Young Investors with Avion Gray, Co-Founder & CEO of Belong 31:59

2M ago31:59

31:59

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Avion Gray, co-founder and CEO of Belong. As an investment platform that uses open banking and AI to assess customers for loans, Belong aims to make long-term investment more transparent and accessible for young people with limited capital. Tune i…

1
S3 Ep54: Empowering Startups for Climate Action with Elliot Coad, Founder of Ecologi 38:11

2M ago38:11

38:11

In this week’s episode of Tech Can’t Save Us, host Paul David is joined by Elliot Coad, Founder and former CEO of Ecologi. As an all-in-one climate platform that aims to help SMBs reduce their carbon footprint and to fund climate projects, Ecologi’s projects range from reforestation in Brazil and providing fuel-efficient cooking stoves in Uganda to…

1
Illustrating Reinforcement Learning from Human Feedback (RLHF) 22:32

2M ago22:32

22:32

This more technical article explains the motivations for a system like RLHF, and adds additional concrete details as to how the RLHF approach is applied to neural networks. While reading, consider which parts of the technical implementation correspond to the 'values coach' and 'coherence coach' from the previous video. A podcast by BlueDot Impact. …

1
Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback 32:19

2M ago32:19

32:19

This paper explains Anthropic’s constitutional AI approach, which is largely an extension on RLHF but with AIs replacing human demonstrators and human evaluators. Everything in this paper is relevant to this week's learning objectives, and we recommend you read it in its entirety. It summarises limitations with conventional RLHF, explains the const…

1
Constitutional AI Harmlessness from AI Feedback 1:01:49

2M ago1:01:49

1:01:49

This paper explains Anthropic’s constitutional AI approach, which is largely an extension on RLHF but with AIs replacing human demonstrators and human evaluators. Everything in this paper is relevant to this week's learning objectives, and we recommend you read it in its entirety. It summarises limitations with conventional RLHF, explains the const…

1
S3 Ep53: Optimizing Urban Transportation with Mobility as a Service with Guillermo Campoamor, Founder & CEO of Meep 42:29

2M ago42:29

42:29

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Guillermo Campoamor, Founder & CEO of Meep. Meep provides companies and cities with the data and intelligence tools they need to achieve more efficient, sustainable trips for everyone. Within the app, Meep offers multimodal route planning, visuali…

1
S3 Ep52: The Future of Flexible & Diverse Workplaces in Tech with Molly Johnson-Jones, Co-Founder & CEO of Flexa 39:40

3M ago39:40

39:40

In this week’s episode of Tech Can’t Save Us, hosts Paul David and Maya Dharampal-Hornby are joined by Molly Johnson-Jones—CEO and co-founder of tech start-up, Flexa. Flexa is revolutionizing the way we work by helping flexible employers get discovered and boost their employer value proposition (EVP) and employer brand. Molly co-founded Flexa to pa…

1
S3 Ep51: Improving Patient Outcomes through AI & NLP with Dipanwita Das, Co-Founder & CEO of Sorcero 37:46

3M ago37:46

37:46

In this week’s episode of Tech Can’t Save Us, host Paul David is joined by Dipanwita Das—CEO and co-founder of tech start-up, Sorcero. Sorcero is revolutionizing the world by utilizing AI to analyze and transform complex medical data into actionable insights. The model also uses natural language processing to help with this, and Sorcero has been aw…

1
S3 Ep50: Insurtech That Prevents Caregiver Burnout with Antonio Ribeiro, Founder & CEO of Yurtle 45:03

3M ago45:03

45:03

In this episode of Tech Can’t Save Us, our guest is Antonio Ribeiro, CEO and founder of Yurtle, the world’s first insurtech company and care management platform for employees balancing work and unpaid caregiving for a loved one. According to the 2021 UK Census, there were 5.7 million unpaid carers in the UK, or 9% of the population. 2.5 million of …

1
S3 Ep49: Why We Need More Trauma-Informed Tech with Devon Geary 33:30

3M ago33:30

33:30

This week’s episode of Tech Can’t Save Us sees Paul speaking to Devon Geary, a senior project manager of a membership organization, to discuss trauma-informed tech, what it is, and why we need to consider it. Devon moved to the UK on a Tech Nation visa and worked here for three years at Birmingham Tech Week. She is an expert in trauma-informed tech…

1
Ep 14 - Interp, latent robustness, RLHF limitations w/ Stephen Casper (PhD AI researcher, MIT) 2:42:17

3M ago2:42:17

2:42:17

We speak with Stephen Casper, or "Cas" as his friends call him. Cas is a PhD student at MIT in the Computer Science (EECS) department, in the Algorithmic Alignment Group advised by Prof Dylan Hadfield-Menell. Formerly, he worked with the Harvard Kreiman Lab and the Center for Human-Compatible AI (CHAI) at Berkeley. His work focuses on better unders…

1
Ep 13 - AI researchers expect AGI sooner w/ Katja Grace (Co-founder & Lead Researcher, AI Impacts) 1:20:28

3M ago1:20:28

1:20:28

We speak with Katja Grace. Katja is the co-founder and lead researcher at AI Impacts, a research group trying to answer key questions about the future of AI — when certain capabilities will arise, what will AI look like, how it will all go for humanity. We talk to Katja about: * How AI Impacts latest rigorous survey of leading AI researchers shows …

1
Deep Double Descent 8:27

3M ago8:27

8:27

We show that the double descent phenomenon occurs in CNNs, ResNets, and transformers: performance first improves, then gets worse, and then improves again with increasing model size, data size, or training time. This effect is often avoided through careful regularization. While this behavior appears to be fairly universal, we don’t yet fully unders…

1
Chinchilla’s Wild Implications 24:57

3M ago24:57

24:57

This post is about language model scaling laws, specifically the laws derived in the DeepMind paper that introduced Chinchilla. The paper came out a few months ago, and has been discussed a lot, but some of its implications deserve more explicit notice in my opinion. In particular: Data, not size, is the currently active constraint on language mode…

1
Eliciting Latent Knowledge 1:00:27

3M ago1:00:27

1:00:27

In this post, we’ll present ARC’s approach to an open problem we think is central to aligning powerful machine learning (ML) systems: Suppose we train a model to predict what the future will look like according to cameras and other sensors. We then use planning algorithms to find a sequence of actions that lead to predicted futures that look good t…

1
Intro to Brain-Like-AGI Safety 1:02:10

3M ago1:02:10

1:02:10

(Sections 3.1-3.4, 6.1-6.2, and 7.1-7.5) Suppose we someday build an Artificial General Intelligence algorithm using similar principles of learning and cognition as the human brain. How would we use such an algorithm safely? I will argue that this is an open technical problem, and my goal in this post series is to bring readers with no prior knowle…

1
Toy Models of Superposition 41:43

3M ago41:43

41:43

It would be very convenient if the individual neurons of artificial neural networks corresponded to cleanly interpretable features of the input. For example, in an “ideal” ImageNet classifier, each neuron would fire only in the presence of a specific visual feature, such as the color red, a left-facing curve, or a dog snout. Empirically, in models …

1
Discovering Latent Knowledge in Language Models Without Supervision 37:09

3M ago37:09

37:09

Abstract: Existing techniques for training language models can be misaligned with the truth: if we train models with imitation learning, they may reproduce errors that humans make; if we train them to generate text that humans rate highly, they may output errors that human evaluators can't detect. We propose circumventing this issue by directly fin…

1
Gradient Hacking: Definitions and Examples 9:15

3M ago9:15

9:15

Gradient hacking is a hypothesized phenomenon where: A model has knowledge about possible training trajectories which isn’t being used by its training algorithms when choosing updates (such as knowledge about non-local features of its loss landscape which aren’t taken into account by local optimization algorithms). The model uses that knowledge to …

1
Two-Turn Debate Doesn’t Help Humans Answer Hard Reading Comprehension Questions 16:39

3M ago16:39

16:39

Using hard multiple-choice reading comprehension questions as a testbed, we assess whether presenting humans with arguments for two competing answer options, where one is correct and the other is incorrect, allows human judges to perform more accurately, even when one of the arguments is unreliable and deceptive. If this is helpful, we may be able …

1
Least-To-Most Prompting Enables Complex Reasoning in Large Language Models 16:08

3M ago16:08

16:08

Chain-of-thought prompting has demonstrated remarkable performance on various natural language reasoning tasks. However, it tends to perform poorly on tasks which requires solving problems harder than the exemplars shown in the prompts. To overcome this challenge of easy-to-hard generalization, we propose a novel prompting strategy, least-to-most p…

1
An Investigation of Model-Free Planning 8:11

3M ago8:11

8:11

The field of reinforcement learning (RL) is facing increasingly challenging domains with combinatorial complexity. For an RL agent to address these challenges, it is essential that it can plan effectively. Prior work has typically utilized an explicit model of the environment, combined with a specific planning algorithm (such as tree search). More …

1
Imitative Generalisation (AKA ‘Learning the Prior’) 18:14

3M ago18:14

18:14

This post tries to explain a simplified version of Paul Christiano’s mechanism introduced here, (referred to there as ‘Learning the Prior’) and explain why a mechanism like this potentially addresses some of the safety problems with naïve approaches. First we’ll go through a simple example in a familiar domain, then explain the problems with the ex…

1
Low-Stakes Alignment 13:56

3M ago13:56

13:56

Right now I’m working on finding a good objective to optimize with ML, rather than trying to make sure our models are robustly optimizing that objective. (This is roughly “outer alignment.”) That’s pretty vague, and it’s not obvious whether “find a good objective” is a meaningful goal rather than being inherently confused or sweeping key distinctio…

1
Empirical Findings Generalize Surprisingly Far 11:32

3M ago11:32

11:32

Previously, I argued that emergent phenomena in machine learning mean that we can’t rely on current trends to predict what the future of ML will be like. In this post, I will argue that despite this, empirical findings often do generalize very far, including across “phase transitions” caused by emergent behavior. This might seem like a contradictio…

1
ABS: Scanning Neural Networks for Back-Doors by Artificial Brain Stimulation 16:08

3M ago16:08

16:08

This paper presents a technique to scan neural network based AI models to determine if they are trojaned. Pre-trained AI models may contain back-doors that are injected through training or by transforming inner neuron weights. These trojaned models operate normally when regular inputs are provided, and mis-classify to a specific output label when t…

1
S3 Ep48: Tech for Worker Empowerment with Nat Whalley, Co-Founder & CEO of Organise 33:31

4M ago33:31

33:31

On this episode of Tech Can’t Save Us, host Paul David welcomes Nat Whalley, co-founder and CEO of Organise, a startup that empowers workers to shape the future of work in the UK and US. Before founding Organise, Nat worked for grassroots organisations like 38 Degrees and Avaaz and led digital strategies at Save the Children. In March 2024, she was…

1
Compute Trends Across Three Eras of Machine Learning 13:50

4M ago13:50

13:50

This article explains key drivers of AI progress, explains how compute is calculated, as well as looks at how the amount of compute used to train AI models has increased significantly in recent years. Original text: https://epochai.org/blog/compute-trends Author(s): Jaime Sevilla, Lennart Heim, Anson Ho, Tamay Besiroglu, Marius Hobbhahn, Pablo Vill…

1
S3 Ep47: The Future of Climate Tech with Anastasia Pavlovic, CEO of Eion 28:30

4M ago28:30

28:30

In this week’s episode of Tech Can’t Save Us, host Paul David is joined by Anastasia Pavlovic—CEO of climate tech company, Eion. After completing her studies in electrical engineering and biosystems, Anastasia started a career in agriculture and climate, becoming a woman in STEM. Since stepping into the role as CEO, Eion has been named one of TIME’…

1
S3 Ep46: Revolutionizing Digital Fundraising with Zaynab Tariq, Co-Founder & COO of CerealBox 34:03

4M ago34:03

34:03

In this episode of Tech Can’t Save Us, host Paul David welcomes Zaynab Tariq, Co-Founder & COO of CerealBox to the show. After working in the nonprofit sector, Zaynab realised that the sector was lagging well behind in terms of tech adoption—and set out on a mission to change it. CerealBox combines growth hacking strategies from marketing with rese…

1
Worst-Case Thinking in AI Alignment 11:35

4M ago11:35

11:35

Alternative title: “When should you assume that what could go wrong, will go wrong?” Thanks to Mary Phuong and Ryan Greenblatt for helpful suggestions and discussion, and Akash Wasil for some edits. In discussions of AI safety, people often propose the assumption that something goes as badly as possible. Eliezer Yudkowsky in particular has argued f…

1
S3 Ep45: Home Energy Efficiency with Robin Peters, Co-Founder & CEO of Snugg 24:37

4M ago24:37

24:37

In this episode of Tech Can’t Save Us, host Paul David is joined by Robin Peters, co-founder and CEO of Snugg, a fintech that focuses on home energy efficiency. Robin is an engineer and entrepreneur who has led and founded several businesses, including deadlyne, which helps businesses meet the UN’s Sustainable Development Goals. Before that, he wor…

1
S3 Ep44: Building Well-Connected Neighborhoods w/ Tech with Michael Wood-Lewis, Co-Founder & CEO of Front Porch Forum 27:53

5M ago27:53

27:53

In this episode of Tech Can’t Save Us, host Paul David is joined by Michael Wood Lewis, Co-Founder & CEO of Front Porch Forum (FPF), an online community-building service in Vermont. Launched in 2006, FPF aims to connect neighbors through local forums that only local people can access. Tune in to hear Paul and Michael discuss the benefits of local o…

1
Public by Default: How We Manage Information Visibility at Get on Board 9:50

5M ago9:50

9:50

I’ve been obsessed with managing information, and communications in a remote team since Get on Board started growing. Reducing the bus factor is a primary motivation — but another just as important is diminishing reliance on synchronicity. When what I know is documented and accessible to others, I’m less likely to be a bottleneck for anyone else in…

1
How to Get Feedback 7:30

5M ago7:30

7:30

Feedback is essential for learning. Whether you’re studying for a test, trying to improve in your work or want to master a difficult skill, you need feedback. The challenge is that feedback can often be hard to get. Worse, if you get bad feedback, you may end up worse than before. Original text: https://www.scotthyoung.com/blog/2019/01/24/how-to-ge…

1
Writing, Briefly 3:09

5M ago3:09

3:09

(In the process of answering an email, I accidentally wrote a tiny essay about writing. I usually spend weeks on an essay. This one took 67 minutes—23 of writing, and 44 of rewriting.) Original text: https://paulgraham.com/writing44.html Author: Paul Graham A podcast by BlueDot Impact. Learn more on the AI Safety Fundamentals website.…

1
S3 Ep43: Generative AI for in-depth qualitative research with Ross Mitchell, Co-Founder of ResearchGOAT 28:48

5M ago28:48

28:48

Ross Mitchell is the CEO and founder of ResearchGOAT, an AI-powered qualitative research platform that helps businesses understand their customers and audiences. He’s also the creative leader at NEOL, a London-based company that helps businesses hire creative leaders for fractional support and assemble project teams quickly. Ross has worked with cl…

1
Being the (Pareto) Best in the World 6:46

5M ago6:46

6:46

This introduces the concept of Pareto frontiers. The top comment by Rob Miles also ties it to comparative advantage. While reading, consider what Pareto frontiers your project could place you on. Original text: https://www.lesswrong.com/posts/XvN2QQpKTuEzgkZHY/being-the-pareto-best-in-the-world Author: John Wentworth A podcast by BlueDot Impact. Le…

1
S3 Ep42: Coaching Tech Startup Founders with Simon Court, Founder of Value Partnership & Author of “Founder’s Legacy” 53:11

5M ago53:11

53:11

In this episode of Tech Can’t Save Us, we welcome Simon Court—coach, founder of Value Partnership (a leadership consultancy firm), and author of ‘Founder’s Legacy: Fifty Game-Changing Leadership Lessons for Building a Great Business’. After working as a management consultant at YSC and HR Director at Ericsson, Simon became a leadership coach and co…

1
S3 Ep41: AI Climate Tech for Industrial Decarbonisation with Dominic Shales, Founder & CEO of Reset Media & DUNE Communications 31:37

5M ago31:37

31:37

Dominic Shales is the founder and CEO of not one but two purpose driven companies: Reset Media, a digital publisher, and DUNE Communications, a corporate comms agency. He also acts as fractional CMO for Nexus Climate, a business that creates and nurtures climate startups, and Carbon Re, an AI climate tech pioneering industrial decarbonisation. In t…

1
How to Succeed as an Early-Stage Researcher: The “Lean Startup” Approach 15:16

5M ago15:16

15:16

I am approaching the end of my AI governance PhD, and I’ve spent about 2.5 years as a researcher at FHI. During that time, I’ve learnt a lot about the formula for successful early-career research. This post summarises my advice for people in the first couple of years. Research is really hard, and I want people to avoid the mistakes I’ve made. Origi…

1
S3 Ep40: DEI, AI and Diversity Washing with Dr. Sarah Saska, Co-Founder & CEO of Feminuity 29:50

5M ago29:50

29:50

Through an evidence-based, data-informed approach paired with leading practices, Feminuity has helped more than 100 companies worldwide build stronger, more inclusive organizations. In this episode of Tech Can’t Save Us, Dr. Saska joins Paul and Wull to discuss the importance of DEI in the tech industry, the threat AI algorithmic bias poses to DEI …

Podcast Berbaloi untuk Didengar

Podcast Bluedot Impact

Podcast Berbaloi untuk Didengar

Panduan Rujukan Pantas