The AI Artifacts Podcast

Brian Warmoth and Sarah Luger
  • 5.0 (5)
  • Technology
  • Updated Weekly

Coming Fall 2023: The AI Artifacts Podcast, co-hosted by Brian Warmoth and Sarah Luger, features timely conversations with founders, technical experts, and practitioners to de-mystify AI's current rise, what it does best, and how it will shape the future. www.aiartifacts.net

  1. 02/04/2024

    Episode 14: Chris Emezue of Lanfrica on AI for low-resource languages

    Researcher and Lanfrica founder Chris Emezue joins the season finale of the podcast to discuss the potential for AI to improve translation across low-resource languages — especially in Africa. Brian and Sarah talk about what’s next for the podcast, as well as where AI has been showing up in earnings reports for GAMMA companies over the past week. Listen for more on those stories and a content-themed edition of "Two truths and l’AI" in this week’s episode. Timestamps for this episode: [0:00] Intro [0:16] Welcome [0:35] What’s next for the podcast [4:42] Tech earnings reports [7:48] Amazon’s new AI shopping assistant Rufus [13:02] Google adds text-to-image functionality to Bard [16:58] Meta’s increased spending on AI [19:58] "Two truths and l’AI" [28:15] Chris Emezue interview [30:27] How he began working with the Bambara language [33:49] His experience in Nigeria and working with other African languages [40:00] Meeting collaborators through Masakhane [44:58] Founding Lanfrica [50:40] What a focus on data farming means [52:57] What constitutes good data when working with languages? [1:04:46] How other people can get involved with AI and language projects Links for topics referenced in this episode: Lanfrica: https://lanfrica.com/ Data farming vs. data mining (Lanfrica): https://lanfrica.com/blog/data-farming-in-the-global-south/ Lanfrica Talks: https://lanfrica.com/blog/lanfrica-talks/ Masakhane website: https://www.masakhane.io/ Microsoft and Alphabet’s recent earnings (The Byte): https://futurism.com/the-byte/ai-companies-lose-190-billion-dismal-google-report Amazon’s new AI shopping assistant (TechCrunch): https://techcrunch.com/2024/02/01/amazon-debuts-rufus-an-ai-shopping-assistant-in-its-mobile-app/ Google adds functionality to Bard (The Verge): https://www.theverge.com/2024/2/1/24057438/bard-gemini-imagen-google-ai-image-generation Zuckerberg’s plan to beat Google and Microsoft in AI (Fortune): https://finance.yahoo.com/news/mark-zuckerberg-explained-meta-crush-004732591.html The AI Artifacts Podcast’s visual branding is by Skaaren Design. Music used in this podcast comes from "Vanishing Horizon" by Jason Shaw and is licensed under an Attribution 3.0 United States License. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.aiartifacts.net

  2. 01/28/2024

    Episode 13: Author Sarah Jaquette Ray on dealing with AI- and climate-related stress

    Sarah Jaquette Ray, a professor at Cal Poly Humboldt, joins Sarah and Brian to talk about dealing with anxiety. As the author of "A Field Guide to Climate Anxiety: How to Keep Your Cool on a Warming Planet,” she discusses some takeaways from her work that can be useful in other contexts, including AI, where major changes are already in progress across industries and careers. In news, several follow-up stories show up as the Federal Trade Commission takes a look at OpenAI and Anthopric’s deals with big tech companies, and Satya Nadella gets asked to weigh in on a wave of Taylor Swift deepfakes on social media. All of this and a new, suspenseful round of "Two truths and l’AI" are in this week’s episode. Timestamps for this episode: [0:00] Intro [0:16] Welcome [1:44] FTC investigation into OpenAI and Anthropic deals [07:16] Microsoft CEO Satya Nadella on Taylor Swift deepfakes [14:30] George Carlin’s estate sues over AI-created comedy special [16:55] AI-created news articles appearing in Google News [24:55] GitHub Copilot study [25:35] "Two truth and l’AI" [36:21] Interview with Sarah Jaquette Ray [40:00] The role of anxiety in climate change and risk [43:30] Moving past fear [51:00] Dealing with burnout and identifying goals [55:04] How to focus attention on specific problems [1:00:00] Confronting negativity bias [1:02:05] Taking a systems thinking approach to mindset Links for topics referenced in this episode: "A Field Guide to Climate Anxiety: How to Keep Your Cool on a Warming Planet" by Sarah Jaquette Ray (University of California Press): https://www.ucpress.edu/book/9780520343306/a-field-guide-to-climate-anxiety "The Existential Toolkit for Climate Justice Educators" edited by by Jennifer Atkinson and Sarah Jaquette Ray (University of California Press): https://www.ucpress.edu/book/9780520397125/the-existential-toolkit-for-climate-justice-educators FTC investigates OpenAI and Anthropic deals (The New York Times): https://www.nytimes.com/2024/01/25/technology/ftc-ai-microsoft-amazon-google.html Satya Nadella on implications of Taylor Swift deepfakes (NBC News): https://www.nbcnews.com/tech/tech-news/taylor-swift-nude-deepfake-ai-photos-images-rcna135913 The Liar’s Dividend (Georgia Tech University): https://gvu.gatech.edu/research/projects/liars-dividend-impact-deepfakes-and-fake-news-politician-support-and-trust-media George Carlin’s estate sues over AI comedy special (ABC News): https://abcnews.go.com/Business/wireStory/george-carlin-estate-sues-fake-comedy-special-purportedly-106700248 AI content appearing in Google News (New York Post): https://nypost.com/2024/01/22/business/google-news-searches-ranked-ai-generated-ripoffs-above-real-articles-including-a-post-exclusive/ GitHub Copilot study (Visual Studio Magazine): https://visualstudiomagazine.com/articles/2024/01/25/copilot-research.aspx AI-generated horse calendar (Bored Panda): https://www.boredpanda.com/ai-generated-horse-calendar/ Volar dating app feature (Wired): https://www.wired.com/story/volar-dating-app-chatbot-screen-matches/ The rise of AI anxiety (Scientific American): https://www.scientificamerican.com/article/ai-anxiety-is-on-the-rise-heres-how-to-manage-it/ Ways the UN believes AI can have a positive impacted on climate change (United Nations): https://news.un.org/en/story/2023/11/1143187 The AI Artifacts Podcast’s visual branding is by Skaaren Design. Music used in this podcast comes from "Vanishing Horizon" by Jason Shaw and is licensed under an Attribution 3.0 United States License. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.aiartifacts.net

  3. 01/21/2024

    Episode 12: Which big tech company is best positioned in AI?

    Alphabet, Amazon, Meta, Apple, and Microsoft each entered 2024 with their own AI strategies, strengths, and weaknesses. In this special episode, Brian and Sarah break down why each company is in a unique position and discuss which of the former GAFAM — now GAMMA (or GAMAM) — is in the strongest place. In news, attention turns to Samsung’s Galaxy S24 announcements, controversy about AI-generated content in Google News results, and a debate between the Justice Department and Federal Trade Commission over where the authority lies to investigate OpenAI. Dive in for all of this an intense crime-themed round of "Two truths and l’AI." Timestamps for this episode: [0:00] Intro [00:16] Welcome and news [1:18] GAMMA adopted to our podcast style guide [2:00] AI-generated face challenge [4:18] Samsung adds AI features to new Galaxy S24 phone [17:43] AI content in Google News results [21:21] Justice Department and FTC discuss jurisdiction over OpenAI [27:28] "Two truths and l’AI" [38:29] 10,000 hours of language recordings to train an AI model [40:14] GAMMA special: What’s going on in AI across the five big tech companies [42:26] Alphabet/Google [50:05] Amazon [54:11] Meta/Facebook [1:02:50] Apple [1:11:48] Microsoft [1:17:56] Who is best positioned, and who is most vulnerable in 2024? Links for topics referenced in this episode: AI-generated face quiz (The New York Times): https://www.nytimes.com/interactive/2024/01/19/technology/artificial-intelligence-image-generators-faces-quiz.html Samsung’s new Galaxy S24 AI features (Wall Street Journal): https://www.wsj.com/tech/samsung-galaxy-s24-artificial-intelligence-features-dc37e134 Samsung’s Galaxy S24 demo video (Samsung): https://www.sammobile.com/news/galaxy-ai-features-video-galaxy-s24-ultra/ Apple passes Samsung as top seller of phones globally (The Guardian): https://www.theguardian.com/technology/2024/jan/17/apple-samsung-worlds-top-smartphone-seller-iphone-honor-google-divide-android Former top seller of 2010, Nokia (Statista): https://www.statista.com/statistics/268877/worldwide-market-share-of-leading-cell-phone-manufacturers/ Assertion that AI-generated content is appearing in Google News results (404 Media): https://www.404media.co/google-news-is-boosting-garbage-ai-generated-articles/ Google’s response to 404 Media story (Search Engine Roundtable): https://www.seroundtable.com/google-responds-garbage-ai-content-in-google-news-36757.html DOJ and FTC meet over jurisdiction to investigate OpenAI (Politico): https://www.politico.com/news/2024/01/19/doj-ftc-microsoft-openai-antitrust-00136624 The AI Taylor Swift-impersonation Le Creuset scam (Creative Bloq): https://www.creativebloq.com/news/taylor-swift-le-creuset-ai Criminals hacking children’s social media accounts (Daily Mail): https://www.dailymail.co.uk/news/article-12859675/Warning-criminals-hacking-childrens-social-media-using-just-three-seconds-voice-turning-terrifying-AI-scam-trick-parents.html How many hours of language recording it takes to train an AI model (LinkedIn): https://www.linkedin.com/feed/update/urn:li:activity:7153882887723884546/ Google anticipates more layoffs in 2024 (CBS News): https://www.cbsnews.com/news/google-layoffs-2024-sundar-pichai/ Microsoft gain cloud share in Q3 2023 (Reuters): https://www.reuters.com/technology/microsofts-ai-bets-boost-cloud-business-alphabet-yet-find-silver-lining-2023-10-24/ Amazon losing $10 billion on Alex in 2022 (Fox Business): https://www.foxbusiness.com/technology/amazon-alexa-track-lose-10-billion-year-described-colossal-failure-new-report# Mark Zuckerberg’s Facebook video about AI (Facebook): https://www.facebook.com/zuck/posts/pfbid02Uf3JSi5xzYqbpEcgfCSmFhwJS3VfrSH5RRxN5C6psFVUWCJADDic8VEADJwPFkVNl Apple’s AI initiatives (Mac Rumors): https://www.macrumors.com/guide/apple-gpt/ Microsoft overtakes Apple as most valuable company (Reuters): https://www.reuters.com/technology/microsoft-overtakes-apple-worlds-most-valuable-company-2024-01-11/ Microsoft alleges spying by state-sponsored Russian group (Reuters): https://www.reuters.com/technology/cybersecurity/microsoft-says-it-was-hacked-by-russian-state-sponsored-group-2024-01-19/ The AI Artifacts Podcast’s visual branding is by Skaaren Design. Music used in this podcast comes from "Vanishing Horizon" by Jason Shaw and is licensed under an Attribution 3.0 United States License. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.aiartifacts.net

  4. 01/14/2024

    Episode 11: David Talbot of Inceptive on generative AI use in biotech

    Brian and Sarah talk to David Talbot, part of the founding team at the startup Inceptive, which is using artificial intelligence to synthesize RNA molecules. It’s a fascinating discussion about AI’s use in biotech and how these two important fields are coming together to introduce new possibilities. In news, enterprise AI use comes up again, as we get some new numbers from OpenAI detailing how many business seats it is serving. In addition, the recap addresses state action on AI regulation and a reaction from the late George Carlin’s daughter to an AI-generated comedy special that used a synthetic voice likeness of her father. Timestamps for this episode: [0:00] Intro [00:17] Welcome and news recap [00:33] OpenAI’s enterprise clients [1:44] State-level action on AI [3:00] AI-generated George Carlin comedy special [6:10] Interview with David Talbot of Inceptive [7:11] David’s career leading up to Inceptive [9:27] AI’s role in designing RNA molecules at Inceptive [11:44] AI’s strengths in this field [14:58] The problems Inceptive is trying to solve [20:18] What kinds of data Inceptive generates [22:54] How machine learning can be used in science [28:00] Watermarking and issues with navigating synthetic data when training AI [30:53] Finding/generating the best data for training [32:34] How biotech has changed how he views AI [34:53] Applying natural language processing competencies in biotech [36:33] What “biological software” means [40:16] Additional details about Inceptive Links for topics referenced in this episode: David Talbot (LinkedIn): https://www.linkedin.com/in/david-t-271a9139/ Inceptive: https://inceptive.life/ OpenAI reveals volume of enterprise clients (Venture Beat): https://venturebeat.com/ai/openai-reveals-how-many-chatgpt-for-enterprise-customers-it-has-so-far/ State-level AI legislation in the U.S. (The New York Times): https://www.nytimes.com/2024/01/11/us/ai-election-ads-state-legislators.html George Carlin’s daughter criticizes AI-generated comedy special (Polygon): https://www.polygon.com/24034625/george-carlin-i-wish-i-was-dead-ai-comedy-special-kelly-carlin-dudes The AI Artifacts Podcast’s visual branding is by Skaaren Design. Music used in this podcast comes from "Vanishing Horizon" by Jason Shaw and is licensed under an Attribution 3.0 United States License. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.aiartifacts.net

  5. 01/07/2024

    Episode 10: Karin Golde on AI and data best practices

    To kick off 2024, Brian and Sarah welcome guest Karin Golde, the founder of West Valley AI and an AWS alum to talk about how she advises companies getting started when they deploy artificial intelligence. What are best practices? What should be avoided? And what needs to be defined going in? In this week’s news, funding gets the spotlight as Perplexity raises $73.6 million and Anthropic’s reported $18.4 billion valuation raises some eyebrows. Then, the conversation turns to U.S. Supreme Court Chief Justice John Roberts’ opinions on AI and how it should be treated in the legal world. Tune in for all of this, along with a CES 2024 edition of "Two truths and ‘AI." Timestamps for this episode: [0:00] Intro [00:17] Welcome [0:58] The FTC’s prize for fighting voice-cloning fraud [3:02] Perplexity raising $73.6 million [10:18] Anthropic reportedly raising $750 million [17:46] Chief Justice Roberts on AI [24:37] "Two truths and l’AI" [33:03] Interview with Karin Golde [34:29] Karin’s experience starting a company [35:48] What to watch out for with AI and data implementations [37:51] Chevrolet of Watsonville’s ChatGPT-powered chatbot incident [43:30] Having a strategy vs. throwing things at the wall to see what sticks [51:32] What types of questions do companies working with Karin ask? [53:40] What should big tech companies know about these non-tech companies adopting AI? [55:15] How to tell if a product is going to solve a problem [1:02:38] What should companies consider when it comes to regulation and long-term thinking? [1:14:40] Most interesting questions about AI going into 2024 Links for topics referenced in this episode: Karin Golde (LinkedIn): https://www.linkedin.com/in/karin-golde/ West Valley AI: https://www.westvalley.ai/ FTC offering $25,000 for solution to fighting voice-cloning fraud (The Register): https://www.theregister.com/2024/01/05/ftc_voice_cloning_solution/ Context page for entries open until January 12 (FTC): https://www.ftc.gov/news-events/contests/ftc-voice-cloning-challenge Perplexity raising $73.6M values it at $520M (Reuters): https://www.reuters.com/technology/perplexity-ai-valued-520-mln-funding-bezos-nvidia-2024-01-04/ Why Anthropic doesn’t want a higher valuation (Semafor): https://www.semafor.com/article/01/05/2024/why-hot-ai-startup-anthropic-wanted-a-lower-valuation Previous reporting on Anthropic’s valuation (Reuters): https://www.reuters.com/markets/deals/anthropic-raise-750-mln-menlo-ventures-led-funding-round-information-2023-12-21/ Chief Justice Roberts weights in on AI (NPR): https://www.npr.org/2024/01/01/1222378766/chief-justice-roberts-casts-a-wary-eye-on-artificial-intelligence-in-the-courts Machine bias in sentencing (ProPublica): https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing Problems with predictive algorithms in policing (MIT Technology Review): https://www.technologyreview.com/2020/07/17/1005396/predictive-policing-algorithms-racist-dismantled-machine-learning-bias-criminal-justice/ Wehead’s ChatGPT edition (Creative Bloq): https://www.creativebloq.com/news/chatgpt-face-wehead LG’s new robot (Yahoo! Finance): https://ca.finance.yahoo.com/news/lg-releases-robot-hopes-lead-170753978.html "Never Too Small" series (YouTube): LG’s StanbyME (LG): https://www.lg.com/us/lifestyle-products/lg-27art10akpl-lifestyle-screen The Chevrolet of Watsonville AI chatbot goes viral (GM Authority): https://gmauthority.com/blog/2023/12/gm-dealer-chat-bot-agrees-to-sell-2024-chevy-tahoe-for-1/ Human labeling being done by domain experts (The New York Times): https://www.nytimes.com/2023/09/25/technology/chatgpt-rlhf-human-tutors.html Kenya workers training AI (The Guardian): https://www.theguardian.com/technology/2023/aug/02/ai-chatbot-training-human-toll-content-moderator-meta-openai The nature of data labeling complexity (The Verge): https://www.theverge.com/features/23764584/ai-artificial-intelligence-data-notation-labor-scale-surge-remotasks-openai-chatbots The AI Artifacts Podcast’s visual branding is by Skaaren Design. Music used in this podcast comes from "Vanishing Horizon" by Jason Shaw and is licensed under an Attribution 3.0 United States License. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.aiartifacts.net

  6. 12/31/2023

    Episode 9: AI predictions for 2024

    In the final episode of 2023, Brian and Sarah go into forecast mode, laying out their AI predictions for 2024. Will the hype keep growing? What will the next phase of adoption look like? This New Year’s episode gets into all of it. Meanwhile, in the week’s biggest news, the discussion looks at The New York Times lawsuit against OpenAI, Jony Ive’s latest hire for some ChatGPT-powered hardware, and Michael Cohen’s AI-powered legal mistake. Timestamps for this episode: [0:00] Intro [00:16] Welcome [1:32] AI predictions for 2024 [4:47] Sarah’s big prediction [10:04] Brian’s big prediction [17:57] The New York Times suing OpenAI [27:04] Jony Ive’s design firm adds Tang Tan, an iPhone designer from Apple [32:14] Michael Cohen’s trouble with Bard [37:52] "Two truths and l’AI" Links for topics referenced in this episode: Skift’s AI-powered chatbot (Skift): https://ask.skift.com/ The New York Times sues OpenAI (Associated Press): https://apnews.com/article/nyt-new-york-times-openai-microsoft-6ea53a8ad3efa06ee4643b697df0ba57 Potential risks for The New York Times (Techdirt): https://www.techdirt.com/2023/12/28/the-ny-times-lawsuit-against-openai-would-open-up-the-ny-times-to-all-sorts-of-lawsuits-should-it-win/ OpenAI’s negotiations with media companies (The New York Times): https://www.nytimes.com/2023/12/29/business/media/media-openai-chatgpt.html Looks at generation AI output that raise copyright concerns (Marcus on AI): Tang Tan from Apple joins Jony Ive (Engadget): https://www.engadget.com/apples-iphone-designer-is-leaving-to-work-with-jony-ive-and-sam-altman-on-ai-hardware-093028794.html Michael Cohen gets caught using AI citing non-existent legal cases (The New York Times): https://www.nytimes.com/2023/12/29/nyregion/michael-cohen-ai-fake-cases.html Champagne at risk, based on ClimateAI data (Fox News): https://www.foxnews.com/lifestyle/new-years-eve-beverage-could-extinct-climate-change-ai-company-predicts Burger King’s facial recognition use for hangover identification (RetailWire): https://retailwire.com/discussion/burger-king-taps-facial-recognition-for-hangover-promotion/ The AI Artifacts Podcast’s visual branding is by Skaaren Design. Music used in this podcast comes from "Vanishing Horizon" by Jason Shaw and is licensed under an Attribution 3.0 United States License. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.aiartifacts.net

  7. 12/23/2023

    Episode 8: Guest Fabian Baier of Pulze.ai on building an AI startup to work with large companies

    This week’s episode brings another startup founder to the show with Pulze.ai CEO Fabian Baier. He talks with Brian and Sarah about creating a product for enterprise companies and what he has learned about working with multiple large language models in a year when so many debuted. In news, the discussion turns to Rite Aid’s issues using facial recognition technology, Apple’s reported talks with media companies about its own AI, and some potential new valuations for OpenAI and Anthropic. Plus, don’t miss this week’s holiday edition of "Two truths and l’AI." Timestamps for this episode: [0:00] Intro [00:16] Holiday welcome [1:10] Rite Aid’s five-year ban on facial recognition tech [11:18] Apple reportedly discussing deals with news publishers to use content for AI training [15:10] OpenAI and Anthropic reportedly seeking additional funding at new valuations [25:31] "Two truths and l’AI" (Sarah was 5-2 before this episode.) [31:50] Interview with Fabian Baier [33:00] Why he started Pulze.ai [37:07] Figuring out what kind of company he wanted to build [57:13] Choosing early use cases and sectors to focus on [59:15] What enterprise companies are asking for [1:03:30] What’s ahead for Pulze.ai Links for topics referenced in this episode: Fabian Baier (LinkedIn): https://www.linkedin.com/in/fbaier/ Fabian’s email: fabian@pulze.ai Pulse.ai: https://www.pulze.ai/ Listeners can use "exclusiveoffer1k" to get 1,000 credits until end of 2023 on: platform.pulze.ai Rite Aid’s settlement with the Federal Trade Commission over AI us (CNN): https://www.cnn.com/2023/12/20/tech/rite-aid-ai-ftc-settlement/index.html Clearview AI reaches settlement in class-action privacy lawsuit (The Record): https://therecord.media/clearview-ai-class-action-privacy-suit-settles-bipa-illinois Apple’s reported talks with Condé Nast, NBC News, and IAC (The New York Times): https://www.nytimes.com/2023/12/22/technology/apple-ai-news-publishers.html OpenAI trying to raise at $100 billion valuation (Bloomberg): https://www.bloomberg.com/news/articles/2023-12-22/openai-in-talks-to-raise-new-funding-at-100-billion-valuation Anthropic looking to raise $750 million at an $18.4 billion valuation (CNBC): https://www.cnbc.com/2023/12/21/openai-rival-anthropic-in-talks-to-raise-750-million-funding-round.html 2023’s big rounds for AI startups (TechCrunch): https://techcrunch.com/2023/12/05/mega-deals-ai-funding-q3/ Airbnb’s use of AI to combat New Year’s Eve parties (Airbnb): https://news.airbnb.com/en-uk/airbnb-using-ai-to-help-clampdown-on-new-years-eve-parties-globally/ AI for saving deer (Andover Advertiser): https://www.andoveradvertiser.co.uk/news/national/24001665.saving-rudolph-ai-used-deter-deer-railway-tracks-christmas/ Andrej Karpathy’s intro to large language models (YouTube):  Pulze.ai’s community (Pulze.ai): https://www.pulze.ai/community Pulze.ai webinar (YouTube): GPT-4 as Rater (Arxiv.org): https://arxiv.org/abs/2308.02575 The AI Artifacts Podcast’s visual branding is by Skaaren Design. Music used in this podcast comes from "Vanishing Horizon" by Jason Shaw and is licensed under an Attribution 3.0 United States License. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.aiartifacts.net

  8. 12/17/2023

    Episode 7: Guest Kurt Bollacker from MLCommons on how to evaluate AI systems

    The podcast returns this week as Brian and Sarah interview Kurt Bollacker, a director of engineering at MLCommons. Ever wonder how to begin testing an AI model? What standards matter? The discussion spans history and model types, with a quick look at recent work out of Stanford. In this week’s news, we get into OpenAI’s relevance for media and wearables with its new Axel Springer deal and reported talks about working with Snap. Listen in, and play along with this week’s festive game of “Two truths and l’AI.” Plus, thanks and full credit to Cory Skaaren at Skaaren Design for our new visual branding. Timestamps for this episode: [0:00] Intro [0:20] Welcome [0:50] OpenAI developments recap [2:23] Axel Springer deal with OpenAI [15:40] Snap’s reported talks with OpenAI to use its object recognition software for smart glasses [17:49] “Two truths and l’AI” [25:59] Interview with Kurt Bollacker [27:50] Kurt’s journey to working on AI data sets and system evaluation [32:14] What are the best standards for evaluating AI? [38:00] Establishing what level of trust to put in AI output [43:00] AI and machine learning history [46:16] Predictions about the future of large language models [50:05] The future of general-purpose AI [53:35] The value of transparency and explainability [55:30] Reviewing the Foundation Model Transparency Index from the Center for Research of Foundation Models at Stanford University [1:00:46] MLCommons work on AI safety testing [1:05:06] LLMs as “plausibility machines,” not “truth machines” [1:07:42] How should the average person use this information? [1:14:26] How the general public views data use for training AI [1:15:58] Likelihood of a new AI winter soon [1:21:49] The AI Safety working group Links for topics referenced in this episode: Axel Springer reaches deal with OpenAI for ChatGPT to use its news content (Axios): https://www.axios.com/2023/12/13/openai-chatgpt-axel-springer-news-deal OpenAI talking to Snap about object recognition software use in smart glasses (The Information): https://www.theinformation.com/articles/tech-giants-chase-wearable-ai The AI-powered Curio and Grimes toy called “Grok” (New York Post): https://nypost.com/2023/12/15/business/grimes-says-grok-toy-has-nothing-to-do-with-elon-musks-ai-bot/ AI-enabled robot pets helping vets (PetFoodIndustry.com): https://www.petfoodindustry.com/pet-ownership-statistics/article/15660175/aianimated-robot-pets-dont-eat-pet-food-or-want-treats MLCommons: https://mlcommons.org/ Internet Archive: https://archive.org/ Water use when training AI models (Brian Warmoth): https://warmoth.org/2023/09/10/how-much-water-ai-is-going-to-drink/ The Foundation Model Transparency Index (Stanford University): https://crfm.stanford.edu/fmti/ The Holistic Evaluation of Language Models (HELM) paper (arxiv.org): https://arxiv.org/abs/2211.09110 MLCommons announcement about the formation of the AI Safety Working Group (MLCommons): https://mlcommons.org/2023/10/mlcommons-announces-the-formation-of-ai-safety-working-group/ The AI Artifacts Podcast’s visual branding is by Skaaren Design. Music used in this podcast comes from "Vanishing Horizon" by Jason Shaw and is licensed under an Attribution 3.0 United States License. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit www.aiartifacts.net

See All (15)

Ratings & Reviews

5
out of 5
5 Ratings

About

Coming Fall 2023: The AI Artifacts Podcast, co-hosted by Brian Warmoth and Sarah Luger, features timely conversations with founders, technical experts, and practitioners to de-mystify AI's current rise, what it does best, and how it will shape the future. www.aiartifacts.net

Information

  • Creator
    Brian Warmoth and Sarah Luger
  • Years Active
    2023 - 2024
  • Episodes
    15
  • Rating
    Clean
  • Copyright
    © Brian Warmoth
  • Show Website
    The AI Artifacts Podcast