Question 1

Can a well-designed prompt outperform OpenAI's o1 model?

Accepted Answer

_Human-written prompts can still beat o1's automated reasoning processes._

Yes. A well-designed prompt, meaning the human writes the instruction, can, so far, outpace OpenAI's o1. This is noteworthy because o1 is designed specifically for reasoning through an iterative, self-calling process with dedicated steps for iteration, reflection, and feedback loops. The fact that careful human-written prompts can still exceed its performance suggests that the model's reasoning capabilities, while advanced, have not yet surpassed what skilled human prompt engineering can achieve.

— [Sam Altman Says, 'We Know How To Build AGI' — Not So Fast](https://www.forbes.com/sites/lutzfinger/2025/01/07/sam-altman---agi--ai-workforce-in-2025-the-battle-of-tech-giants/) · _Forbes_

Question 2

Does Google still have a chance against the new AI search competitors?

Accepted Answer

_Google's experience and data give it staying power in AI search._

Yes. Don't count Google out just yet. Google has a significant advantage with its years of experience and vast amounts of data. The comeback it made with its generative search experience was impressive. While Google was late to the game, its generative search experience is coming back quite strong. It won't be the end of Google, but the fight is definitely on.

— [2025: The Year Of Search But Not The End Of Google](https://www.forbes.com/sites/lutzfinger/2025/01/03/2025-the-year-of-search-but-not-the-end-of-google/) · _Forbes_

Question 3

How accurate are current deepfake detection technologies?

Accepted Answer

_Detection tools fail more often than they succeed right now._

Current deepfake detection technology isn't very good yet. The best deepfake detector is only 65% accurate. Even Microsoft's Azure Cognitive Services was fooled over 78% of the time. This means detection tools fail more often than they succeed. The future of these detectors will mirror bot detectors, spam detectors, or any other cyber threat detectors. It will be an arms race between those creating deepfakes and those trying to detect them, similar to how spam and virus detection evolved over time.

— [Deepfakes - The Danger Of Artificial Intelligence That We Will Learn To Manage Better](https://www.forbes.com/sites/lutzfinger/2022/09/08/deepfakesthe-danger-of-artificial-intelligence-that-we-will-learn-to-manage-better/) · _Forbes_

Question 4

How are AI agents different from RPA (robotic process automation)?

Accepted Answer

_Agents add reasoning where RPA's rule engine stops._

RPA is a rule engine. It automates structured, low-skilled tasks where every decision is essentially zero or one, which is why UiPath and Automation Anywhere built real businesses but hit a ceiling: a script either breaks or runs, and it cannot handle uncertainty or unstructured data. Agents add a language model on top, so they can read an email's style, cope with ambiguity, and self-correct. Cherry's portfolio company Octomind does this for front-end testing with a self-healing loop that notices when a test goes wrong and tries another path, something a rule engine cannot do. The catch: you still have to give an agent clear boundaries, or it goes rogue.

— [The Edge: “AI Agents: The Future of Work?”](https://www.lutzfinger.com/transcript/2024-12-12-ai-agents-the-future-of-work/) · _Podcast_

Question 5

How can AI support frontline workers who aren't sitting at computers all day?

Accepted Answer

_Think voice-first interfaces and knowledge retrieval, not just desktop tools._

Frontline workers move around and might not have time to sit at a laptop, but they could have a quick second to speak or type a question. You could support them with an AI interface that answers questions in their language, translates into technical knowledge, and searches knowledge databases. The key is making sure answers are right. Don't just put ChatGPT on your FAQs. You need human in the loop at the beginning, testing a lot to ensure frontline workers are happy with the output before rolling it out widely.

— [You asked, we answered](https://www.lutzfinger.com/transcript/2023-07-28-you-asked-we-answered/) · _The Edge_

Question 6

How can startups use large language models to improve their products?

Accepted Answer

_Simplify interfaces and gain user insights through language_

As a startup founder, think about which interface will be changed by language input. Whenever users needed to use their mouse or learn complex clicking and dragging, you suddenly have the ability to let them do it via language. This simplifies the interface dramatically. Additionally, language prompting gives you more feedback. Before, you'd create an interface and see how users clicked buttons and moved around without really knowing their intentions. Now with language interfaces, you actually see what they're searching for and asking for. This gives you valuable input for building your product.

— [AI as an interface](https://www.lutzfinger.com/transcript/2023-04-24-ai-as-an-interface/) · _The Edge_

Question 7

How can transparency tools like NFTs help combat deepfakes?

Accepted Answer

_NFTs could verify authentic videos by tracking their original sources._

In a world of deepfake abundance, verified accounts alone will not be sufficient to instill trust. One proposition to combat deepfakes points to the usage of NFTs. If everyone encrypted their videos as NFTs, it would be easy to find and compare different sources of the same moment. This transparency mechanism would help establish which version of a video is authentic by creating a verifiable chain of custody. Social media companies created verified accounts as one trust mechanism, but new technologies like NFTs may be needed as the deepfake threat grows.

— [Deepfakes - The Danger Of Artificial Intelligence That We Will Learn To Manage Better](https://www.forbes.com/sites/lutzfinger/2022/09/08/deepfakesthe-danger-of-artificial-intelligence-that-we-will-learn-to-manage-better/) · _Forbes_

Question 8

How did the Apple Card algorithm discriminate despite not using gender as a direct feature?

Accepted Answer

_Apple Card offered vastly different limits to spouses with identical assets._

David Heinemeier Hansson was offered a credit limit 20 times higher than his wife despite having the same shared assets. The algorithm likely learned from a world where men earn on average more than women and concluded that men are more creditworthy. This shows how bias creeps in even when protected characteristics like gender aren't direct features. The training data itself was biased, reflecting existing societal inequalities, and the algorithm learned to perpetuate them without explicitly considering gender.

— [It's The Data, Stupid! Why AI Might Get It Wrong.](https://www.forbes.com/sites/lutzfinger/2020/05/28/its-the-data-stupid-why-ai-might-get-it-wrong/) · _Forbes_

Question 9

How did you replace yourself with a bot in your AI and product course?

Accepted Answer

_A virtual instructor keeps course content fresh and current._

I launched a public certificate program on product and AI at Cornell's Johnson School. The course is about 100 hours long. To keep the content up to date, I replaced myself with a bot, a virtual copy of myself. This allows the course material to stay current without requiring me to manually update everything. It's an example of how AI can maintain educational content in a scalable way while I focus on other work, including running my Gen AI platform for e-commerce.

— [Disrupting Traditional Banking, with Colin Walsh](https://www.lutzfinger.com/transcript/2025-05-22-disrupting-traditional-banking-with-colin-walsh/) · _The Keynote on AI_

Question 10

How do chained AI models create better workflows than simple generative AI tools?

Accepted Answer

_Building dynamic workflows by connecting AI models in sequence._

Chained AI models connect multiple models together so the output of one becomes the input for the next, creating dynamic workflows rather than fixed steps. For an email assistant, the system first analyzes the email and figures out needed steps, then checks the calendar, looks up past emails, and predicts likelihood of attending. It creates three draft replies with different options and presents them to the user. This approach enables system integration, context search through RAG, predictions using traditional AI, and user control through multiple options.

— [AI Agents In 2025: What Enterprise Leaders Need To Know](https://www.forbes.com/sites/lutzfinger/2025/01/05/ai-agents-in-2025-what-enterprise-leaders-need-to-know/) · _Forbes_

Question 11

How do encoder-decoder pairs work to create deepfakes?

Accepted Answer

_Extracting and reconstructing facial features to swap identities in images._

Encoder-decoder pairs are one of the main tools for creating deepfakes today. Going from an original image to a latent image is called encoding. Going from latent back to image is called decoding, similar to how you close your eyes and picture a cat based on knowledge stored in your brain. The encoder extracts latent features of face images, and the decoder reconstructs the face images from those features. To generate a deepfake, the decoder for the target draws the target image with the source's latent features, such as expressions.

— [Overview Of How To Create Deepfakes - It's Scarily Simple](https://www.forbes.com/sites/lutzfinger/2022/09/08/overview-of-how-to-create-deepfakesits-scarily-simple/) · _Forbes_

Question 12

How do Generative Adversarial Networks make deepfakes more realistic?

Accepted Answer

_Two competing neural networks train each other to create realistic fakes._

Generative Adversarial Networks, or GANs, were introduced in a 2014 academic study by Goodfellow and colleagues. GANs set up two neural networks to compete against each other. The first network, a generative neural network, creates a realistic image from a random seed through decoding. The second, a discriminative classifier, checks whether the image is real or fake. Through this competition, the two neural networks train each other and become more and more realistic over time. This architecture reinvigorated interest in deepfakes.

— [Overview Of How To Create Deepfakes - It's Scarily Simple](https://www.forbes.com/sites/lutzfinger/2022/09/08/overview-of-how-to-create-deepfakesits-scarily-simple/) · _Forbes_

Question 13

How do you decide if AI is the right fit for your product problem?

Accepted Answer

_Not every problem needs AI; start with whether complexity demands it._

We created an AI fit analysis with portfolio companies. It starts with questions around precision and output quality, focusing on data quality. Then ask: Do you benefit from modeling complex decisions versus just a few? If it's a linear problem, a rule engine works. But if you have a complex decision space or nonlinear problems like search engines or recommendation engines, then it's time to deploy AI to find tiny clues. This doesn't always mean large language models. Sometimes simple regression analysis is good enough and more precise.

— [Building AI products: 5 lessons from our founders' workshop](https://www.lutzfinger.com/transcript/2024-08-01-building-ai-products-5-lessons-from-our-founders-workshop/) · _The Edge_

Question 14

How does AI change healthcare?

Accepted Answer

_AI applications, value-based care, regulation._

> I just got back from the JP Morgan conference. A few learnings that apply to any industry, not just healthcare:Humans + AI will change the industry. Not AI alone.Cash flow matters. Not AI hype.The real AI revolution is operational.One of the most hyperbolic promises in healthcare in the last month was the claim that: "AI is going to design our drugs and will replace researchers." In my view, this is not true.

— [Healthcare's AI Lesson: Autocomplete Isn't Understanding](https://www.forbes.com/sites/lutzfinger/2026/01/16/healthcares-ai-lesson-autocomplete-isnt-understanding/) · _Forbes_

Question 15

How does AI change the way we interact with computers compared to previous interfaces?

Accepted Answer

_From punch cards to conversation: AI as natural interface_

AI enables me to communicate with machines using natural language instead of learning complex commands or navigating graphical interfaces. Previously, we evolved from punch cards to MS-DOS commands to graphical interfaces with mice and clicking. Each step made things more intuitive, but I still had to search and learn the system's rules. Now with large language models, I can simply describe what I want in my own words. The AI understands my intent and produces output, even if I have no idea how it works technically. This is revolutionary because humans naturally communicate through language.

— [AI as an interface](https://www.lutzfinger.com/transcript/2023-04-24-ai-as-an-interface/) · _The Edge_

Question 16

How does AI enable better information retrieval than traditional search?

Accepted Answer

_LLMs transform how we access and summarize information across all domains._

Large language models can read language, extract and summarize knowledge, and create new data based on what you're asking for. The internet democratized data, but data still needs processing and summarizing into chunks we can act on. LLMs make this chunking more effective. It's like how you can watch a two-hour movie and then concisely tell a friend the content. An LLM can do this for you, providing compact, concise information retrieval. This isn't just one size fits all. In education, for example, you might have many explanations for a math problem, and the question is which explanation works best for a specific student.

— [Cherry's Investment Thesis on AI](https://www.lutzfinger.com/transcript/2024-07-19-cherrys-investment-thesis-on-ai/) · _The Edge_

Question 17

How does AI find patterns in data that humans miss, like in healthcare?

Accepted Answer

_AI detects invisible patterns in healthcare data to prevent emergencies._

AI takes data and finds tiny clues that you and I might not see. At Google Health, we worked with electronic health records to predict whether somebody would end up in an emergency room six months before it happened. The AI sees the writing on the wall in the data. Roughly one third of people in hospitals shouldn't be there if we would have just known. The computer gets trained on data and figures out those tiny clues to predict preventable hospitalizations.

— [AI, Clones, Deep Fakes, and Neural Networks](https://www.lutzfinger.com/transcript/2024-09-12-ai-clones-deep-fakes-and-neural-networks/) · _The Keynote on AI_

Question 18

How does Catalog ensure certain workflows always follow the same steps when using LLMs?

Accepted Answer

_A rules engine ensures critical workflows always execute consistently._

We built a workflow engine underneath our LLMs. The LLM generates a workflow on the fly every time, but users can design deterministic journeys inside our system using workflows (like Zapier but internal). If a request matches a workflow definition, it triggers that workflow every time instead of a dynamically generated journey. It's no code, so users can do it themselves. First, the LLM does a workflow lookup to see if there's something you always want to do, then for the rest it goes into free flow logic.

— [Large Language Models: one interface to rule them all?](https://www.lutzfinger.com/transcript/2023-08-15-large-language-models-one-interface-to-rule-them-all/) · _The Edge_

Question 19

How does Catalog prevent large language models from hallucinating when retrieving enterprise information?

Accepted Answer

_Stopping hallucinations by feeding LLMs the right information first._

The way to avoid hallucination is to stop the model from reaching into the model layers for information. If you give it structured or unstructured information in raw format and then ask it to summarize or list, it gives you exactly what you provided. Without that context, it makes things up because it's doing probabilistic retrieval from its training. This is retrieval augmented generation. You get the information first, augment the prompt with it, then generate the response based on that specific context.

— [Large Language Models: one interface to rule them all?](https://www.lutzfinger.com/transcript/2023-08-15-large-language-models-one-interface-to-rule-them-all/) · _The Edge_

Question 20

How does Catalog use large language models to simplify enterprise software access?

Accepted Answer

_One interface to access all your enterprise tools at once._

Catalog uses LLMs to create a unified interface for all enterprise software. Instead of navigating multiple systems, employees enter a single prompt. The LLM understands the query, identifies which data repository holds the answer (like Workday for payslips), converts the request into an API call, performs deterministic retrieval, and returns the result. We don't use LLMs for data retrieval itself, only for reasoning and synthesis. This compresses workflows from multiple surfaces and ten to fifteen clicks down to one query and two or three clicks.

— [Large Language Models: one interface to rule them all?](https://www.lutzfinger.com/transcript/2023-08-15-large-language-models-one-interface-to-rule-them-all/) · _The Edge_

Question 21

How does jamming affect drone operations, and why does it make autonomy necessary?

Accepted Answer

_Communication denial forces drones to decide without waiting for human orders._

Jammers emit white noise across frequencies to drown out communication. GPS satellite signals go first because they are weak. Your connection to home base gets jammed second because you are far away and the jammer is closer. But drones close together, only hundreds of meters apart, can still communicate with each other because the nearest jammer is kilometers away. You have to assume information from home base arrives intermittently at best. This is why drones must make autonomous decisions. They cannot wait for human instructions that may never arrive.

— [AI on the Front Lines, with Alexander Fink](https://www.lutzfinger.com/transcript/2026-04-30-ai-on-the-front-lines-with-alexander-fink/) · _The Keynote on AI_

Question 22

How does LegalOS use LLMs to make legal knowledge more accessible?

Accepted Answer

_LLMs translate legal knowledge into something everyone can use._

LegalOS originally built a no-code visual diagramming tool that lets legal experts encode their knowledge in a structured, machine-readable way. It's like a programming language for law. In recent months, we've pivoted to use LLMs much more heavily to make that legal knowledge accessible. LLMs provide a higher level of abstraction for automations. They let us map open-ended legal requests to structured tasks without requiring someone to manually encode every single rule. The LLM acts as an interface or translator, so humans don't have to understand legalese directly but can still interact with legal text.

— [Will LLMs scale and revolutionize the legal space?](https://www.lutzfinger.com/transcript/2023-06-15-will-llms-scale-and-revolutionize-the-legal-space/) · _The Edge_

Question 23

How does Marpai Connect use AI to overcome the small numbers problem?

Accepted Answer

_Pooling clients and predictive AI create scale for small employers._

Marpai Connect addresses the small numbers challenge by pooling all self-insured clients and anticipating an overall reduction in aggregate. We use various approaches from simple time series modeling to advanced deep learning models like Neural Prophet to predict risk based on longitudinal claims records. This pooling strategy allows us to achieve the statistical power that individual small employers lack, while still delivering personalized care through AI matchmaking that pairs each user with the most suitable value-based care partner.

— [Changing Healthcare - A Platform Approach Using Data & AI](https://www.forbes.com/sites/lutzfinger/2023/07/05/changing-healthcarea-platform-approach-using-data--ai/) · _Forbes_

Question 24

How does multimodal AI improve the healthcare experience for doctors dealing with overwhelming patient data?

Accepted Answer

_Summarizing mountains of patient data to surface what truly matters._

In healthcare, patients often overwhelm doctors with massive amounts of data. For example, a patient with atrial fibrillation might bring five years of sleep data from their smartwatch, or a cancer patient could arrive with a 20-pound stack of medical records. What's needed is an AI that can summarize and highlight the key points. Multimodal AI handles diverse data sources like text, images, audio, video, and time-series data simultaneously, offering a richer, more comprehensive understanding to help doctors focus on what matters most.

— [Multimodal AI In 2025: From Healthcare To eCommerce And Beyond](https://www.forbes.com/sites/lutzfinger/2025/01/06/multimodal-ai-in-2025-from-healthcare-to-ecommerce-and-beyond/) · _Forbes_

Question 25

How does Neuralens measure brain health non-invasively compared to current methods?

Accepted Answer

_Replacing skull drilling with light-based sensors for brain monitoring._

Currently, doctors literally drill a burr hole through your skull with a household drill and insert sensors through brain tissue to measure oxygen saturation, blood flow, and intracranial pressure. This invasive surgery is only done on 200,000 to 300,000 patients yearly, though over 10 million neurocritical care patients could benefit. Neuralens uses advanced methods of measuring how light interferes with brain tissue to quantify blood flow, blood volume, and intracranial pressure non-invasively. We can stick sensors on the head or shine a laser from 30 centimeters away to make these measurements.

— [Founder Coaching: Tips for your AI Journey](https://www.lutzfinger.com/transcript/2024-07-26-founder-coaching-tips-for-your-ai-journey/) · _The Edge_

Question 26

How does Think Labs AI use physics informed AI differently than general purpose models like ChatGPT?

Accepted Answer

_Grid AI must be grounded in physics, not just patterns in data._

We cannot just put SCADA measurements and smart metering information into ChatGPT or Microsoft Copilot or Gemini and ask how the grid works. It does not work like that. We lean into something called Physics Informed AI. We train our AI models using physics. It is physics driven, physics based, and physics accurate. We teach AI how the grid works and reproduce engineering analysis faithfully. The AI can turn traditional studies that take weeks to months into minutes, and go from analyzing one feeder at a time into full system analysis.

— [AI Needs Power, the Grid Needs AI, with Josh Wong](https://www.lutzfinger.com/transcript/2026-04-02-ai-needs-power-the-grid-needs-ai-with-josh-wong/) · _The Keynote on AI_

Question 27

How does TikTok's algorithm differ from Clubhouse's approach to content discovery?

Accepted Answer

_TikTok uses behavior and content analysis, not social graphs._

TikTok is an algorithm-first company. Unlike Clubhouse or Facebook, it does not mainly rely on the social graph but has built a powerful recommendation engine. TikTok's machine learning magic is a supervised learning system that finds videos interesting to the user by measuring how long you look, how fast you scroll, and many other behavioral features. Additionally, the algorithm evaluates the actual video clip: content, speed, sound, colors, words. Bytedance has an absurd proportion of their software engineers focused on their algorithms. This data has created a sizable moat that not even Facebook could cross easily.

— [Clubhouse's Future Depends On Data - How To Build A TikTok Like Algorithm](https://www.forbes.com/sites/lutzfinger/2021/02/09/clubhouses-future-depends-on-datahow-to-build-a-tiktok-like-algorithm/) · _Forbes_

Question 28

How does your company Gooder AI address the gap between technical metrics and business value?

Accepted Answer

_*Translating predictive model performance into profit and savings estimates*_

Gooder addresses one particular place where there's a need for a new category of software: the metrics. It's a business user console for understanding potential value and making decisions about deployment. Typically, data scientists present technical metrics like area under the curve, precision, or recall, but very rarely go to stakeholders and say if you deploy this model, it has potential value of this much profit or savings. We provide visualization where both dimensions are business relevant, like the decision threshold and business metrics like savings or profit, so stakeholders can see exactly what their options are.

— [The Value Translation Gap: AI's Deployment Problem](https://www.lutzfinger.com/transcript/2024-11-28-the-value-translation-gap-ais-deployment-problem/) · _The Edge_

Question 29

How has natural language processing changed since 2015, and why does that matter for legal tech?

Accepted Answer

_From counting words to real understanding in less than a decade._

I quit my PhD in 2015 and co-founded a natural language processing company. Back then, the state of the art was basically counting words in a glorified way. You had an idea of what the user might be talking about, but you couldn't map it to something with certainty. You had to collect your own data and train specialized models to do one narrow task. LLMs changed that. You can use foundational models out of the box without fine-tuning. They solve many tasks already. Understanding text and mapping open-world requests to closed sets of tasks became way more accessible.

— [Will LLMs scale and revolutionize the legal space?](https://www.lutzfinger.com/transcript/2023-06-15-will-llms-scale-and-revolutionize-the-legal-space/) · _The Edge_

Question 30

How has online shopping changed over the past 25 years?

Accepted Answer

_Twenty-five years later, online shopping looks surprisingly familiar._

Surprisingly, e-commerce hasn't changed much in the last 25 years. If you compare Amazon in 1999 to Amazon today, the core structure is essentially the same. You still need to select a category, then a subcategory, and you better know what you want and where it fits. The 1999 version had long written lists. Today looks slicker with more images, but shopping is still complicated. Most of the time, subcategories let you filter by stars or price, but they don't really describe product dimensions. For those details, you still need to click into the product and read if you care. It's still a transactional act of buying.

— [AI in E-Commerce](https://www.lutzfinger.com/transcript/2024-10-03-ai-in-e-commerce/) · _The Keynote on AI_

Question 31

How has the shift from SEO to GEO changed what websites need to do for brands?

Accepted Answer

_Websites now feed AI rather than convert human visitors directly._

Websites used to be designed to convert human visitors, but now their main purpose needs to be feeding data into AI. The three-party model of searcher, Google, and end content is disintegrating into a two-party model. Decisions are getting made before consumers hit the website, so the website is no longer the main conversion point. Instead, websites should be clean data delivery mechanisms to computers, very explicit about who you sell to, your target use case, pain points, and the entire buyer journey. If it's not explicit, the AI has no way to know and might recommend somebody else.

— [Marketing to Humans Changed, with Alex Dees](https://www.lutzfinger.com/transcript/2026-03-24-marketing-to-humans-changed-with-alex-dees/) · _The Keynote on AI_

Question 32

How is OpenAI trying to improve user experience to build stickiness?

Accepted Answer

_Making ChatGPT stickier by reducing friction and frustration for users._

ChatGPT has a moat because over 100 million consumers go to ChatGPT instead of Google with their questions. However, competition is just a click away. LLMs are not friction-free and can easily frustrate users by giving lengthy, irrelevant or incorrect information. ChatGPT started to iterate with its interface to improve the UX, offering pre-written queries and warnings about accuracy. The easier the access to ChatGPT, the stickier the product. OpenAI will continue improving the user experience to maintain this advantage.

— [How OpenAI Can Secure Its Investments - A Playbook That Needs Microsoft](https://www.forbes.com/sites/lutzfinger/2023/08/18/the-potential-moat-of-openaihow-openai-can-secure-its-investments/) · _Forbes_

Question 33

How is r2decide using multimodality to transform the online shopping experience?

Accepted Answer

_Merging search, browse, and chat into one seamless shopping flow._

At r2decide, we're using multimodality to merge Search, Browse, and Chat into one seamless flow. In an online jewelry store, a user searching for "green" would see results based on both text and images, since our AI encodes images into a shared latent space. Users can also search for broader contexts like "wedding," "red dress," or "gothic." This capability even extends to brand names like "Swarovski," surfacing relevant items even if the shop doesn't officially carry Swarovski products.

— [Multimodal AI In 2025: From Healthcare To eCommerce And Beyond](https://www.forbes.com/sites/lutzfinger/2025/01/06/multimodal-ai-in-2025-from-healthcare-to-ecommerce-and-beyond/) · _Forbes_

Question 34

How should product builders think about data bias in AI systems?

Accepted Answer

_Bias in data is inevitable; product decisions determine how much to accept._

There is never a way to have non-biased data, because data is biased. As a product person, you need to decide by how much you want to de-bias the data. Sometimes bias is actually good because you don't want averages, you want the best outcome. For example, if identifying certain patients or optimal machinery output, you don't want what the machine did in the past all the time, but just the best parts of it. You're injecting very good data versus just average data.

— [Building AI products: 5 lessons from our founders' workshop](https://www.lutzfinger.com/transcript/2024-08-01-building-ai-products-5-lessons-from-our-founders-workshop/) · _The Edge_

Question 35

How will Large Language Models impact healthcare according to Lutz?

Accepted Answer

_LLMs will save lives by transforming patient engagement and interaction._

Large Language Models will improve the last mile with patients, which is huge because engaging patients improves outcomes. In other words, LLMs will save lives. The major impact in healthcare is that we're about to change the way human-AI interaction works. LLMs allow a new way to interact with technology, enabling personalized messages, AI-powered chatbots that send tailored advice based on personal preferences, and knowledge base systems that explain diagnoses and educate patients on pathways.

— [Large Language Models & AI In Healthcare](https://www.forbes.com/sites/lutzfinger/2023/03/27/large-language-models--ai-in-healthcare/) · _Forbes_

Question 36

If agents are the future, why won't one giant model just do everything?

Accepted Answer

_The moat is the workflow, not a bigger model._

Because the durable approach is the opposite: smaller, faster, specialized models orchestrated together, not one ever-bigger general model. Even OpenAI orchestrates the long tail of consumer search with smaller models, because a big model would be far too slow. For founders this is the key warning: if your agent's next step is something generic, like deciding whether a calculator is needed, that is the long tail, and the big models will absorb it, just as Google eventually swallowed the specialized search engines. The defensible business is one where the founder knows the specific stepwise workflow of a domain and chains the right models and rules to serve it.

— [The Edge: “AI Agents: The Future of Work?”](https://www.lutzfinger.com/transcript/2024-12-12-ai-agents-the-future-of-work/) · _Podcast_

Question 37

If OpenAI doesn't have a model moat, what kind of moat does it actually have?

Accepted Answer

_User base of 100M offers paths to sustainable competitive advantage_

OpenAI doesn't have a model moat, but it has a 100 million users moat. ChatGPT currently has over 100 million users, which creates opportunities to build sustainable advantages. There's much that can be done to create lasting competitive protection: User Experience, Data Access, Workflow Integration, LLM governance, and regulatory protection. While the business model should give us pause, the user base provides a foundation for building real defensibility. The moat comes from how you serve and retain those users, not from the underlying technology itself.

— [OpenAI Isn't Going Bankrupt, But It Has A Business Model Problem](https://www.forbes.com/sites/lutzfinger/2023/08/18/is-openai-going-bankrupt-no-but-ai-models-dont-create-moats/) · _Forbes_

Question 38

Is ChatGPT actually at risk of shutting down despite the bankruptcy headlines?

Accepted Answer

_Microsoft backing ensures survival despite user decline and cost concerns_

No. ChatGPT has rapidly evolved into a prominent consumer service. No other application has grown at such a pace, and it's here to stay. Microsoft sees value in OpenAI as a tool to compete with Google and will likely continue funding server costs. While OpenAI has seen a drop in users, bankruptcy is off the table. The attention-grabbing headlines don't reflect the reality. ChatGPT isn't going anywhere. The real issue isn't survival but rather the underlying business model, which has concerning weaknesses around creating sustainable competitive advantages.

— [OpenAI Isn't Going Bankrupt, But It Has A Business Model Problem](https://www.forbes.com/sites/lutzfinger/2023/08/18/is-openai-going-bankrupt-no-but-ai-models-dont-create-moats/) · _Forbes_

Question 39

Isn't OpenAI's o1 model a first step toward AGI?

Accepted Answer

_O1's chained reasoning is impressive but still not intelligence._

No. OpenAI's o1 does not directly answer a question. Instead, it creates a plan to determine the best way to answer, then critiques its response, improves it, and continues refining. This chained output is truly impressive. But this is not intelligence. It's a fixed framework and not dynamic. The more critique cycles you allow the model, the more facets it can address, but this comes at the cost of increased time. It's still pattern-based work, not real intelligence.

— [No AGI But A 'Killer App' — 2025 AI Prediction](https://www.forbes.com/sites/lutzfinger/2025/01/03/no-agi-but-a-killer-app2025-ai-prediction-110/) · _Forbes_

Question 40

Should enterprises invest equally in predictive and generative AI, or focus on one?

Accepted Answer

_*Predictive and generative AI should not compete for the same budget*_

Enterprises should really be investing at least as much into the potential value from predictive use cases as generative. Right now it's becoming kind of a zero-sum game, with a lot of budgets moving away from predictive. But predictive is the technology you turn to for improving your existing large-scale operations. These are very different kinds of value propositions. They should compete no more than water parks and ski resorts, but they totally are competing. We need to be doing both. Most of predictive's potential is still on tap.

— [The Value Translation Gap: AI's Deployment Problem](https://www.lutzfinger.com/transcript/2024-11-28-the-value-translation-gap-ais-deployment-problem/) · _The Edge_

Question 41

Should founders focus on building new AI models or developing applications with existing technology?

Accepted Answer

_Applications beat innovation as VCs shift from research to practical tools._

Take the existing capabilities and apply them. Do we really need more innovation, or is it fine to say large language models are very good, you can chain them, combine them with traditional machine learning techniques, and build an application around it? I want more practicality, more tooling around it. The VC world actually sees that the appetite to go on risky, more researchy projects has gone down. VCs want more applications. Founders should focus on taking existing capabilities and applying them rather than pursuing foundational model innovation.

— [AI Gold Rush: Is the Market Overheating?](https://www.lutzfinger.com/transcript/2024-08-08-ai-gold-rush-is-the-market-overheating/) · _The Edge_

Question 42

Should you trust OpenAI blindly?

Accepted Answer

_Vendor risk and architectural diversification._

> Stargate's $500B investment — message to the competitors — meant to intimidate potential challengers with overwhelming financial backing. AI is extremely dangerous; only a select few should be allowed to develop it — OpenAI's storyline to use safety concerns as justification for a regulatory setup that creates barriers to entry and gives them more monopolistic power. From now on, models should respect IP rights — OpenAI ignored IP…

— [Is OpenAI A Victim — DeepSeek And IP Rights — A Playbook For Leaders](https://www.forbes.com/sites/lutzfinger/2025/01/29/is-openai-a-victim---deepseek-and-ip-rights---a-playbook-for-leaders/) · _Forbes_

Question 43

What advantage do existing platforms like Google, Meta, and TikTok have in the bot space?

Accepted Answer

_Existing user access matters, but cross platform integration creates opportunity_

Companies with existing access to users and data will likely have the majority chunk of the market. While we are still figuring out style, entertainment value, and answer quality, TikTok, Google, Meta, X, and Snap are all putting out bots to monetize their existing access. They already have the right information and the right tools. However, there is an opportunity for cross platform bots that operate across different systems, like accessing both SAP and HubSpot, or working across Snap and LinkedIn, following users regardless of platform.

— [The future of AI bots: winning with personalization, conviction, and cross-platform integration](https://www.lutzfinger.com/transcript/2024-08-22-the-future-of-ai-bots-winning-with-personalization-conviction-and-cross-platform/) · _The Edge_

Question 44

What are AI-generated nudges and how do they work in r2decide's system?

Accepted Answer

_Contextual recommendations that guide shoppers toward the best options._

Alongside search results, r2decide also generates AI-driven nudges, which are contextual recommendations designed to guide users effortlessly toward the most relevant options. These nudges provide chat-like advice, helping shoppers make better decisions without having to manually filter through endless product listings. By combining multimodal understanding with these intelligent prompts, the system transforms the traditional eCommerce interface into something more interactive and helpful, moving beyond the keyword-and-scroll experience that has dominated for 25 years.

— [Multimodal AI In 2025: From Healthcare To eCommerce And Beyond](https://www.forbes.com/sites/lutzfinger/2025/01/06/multimodal-ai-in-2025-from-healthcare-to-ecommerce-and-beyond/) · _Forbes_

Question 45

What are the five strategies OpenAI is using to create a business moat?

Accepted Answer

_Five strategies to protect OpenAI's business beyond the model itself._

OpenAI's playbook includes five strategies: First, invest in a better user experience focused on LLMs. Second, create a feedback loop that improves their LLMs over time. Third, integrate into many workflows to improve utilization. Fourth, allow BYOD (bring your own data), since enterprise private data is king. Fifth, ask the regulator to protect you. Many of these potential moats depend on Microsoft. While LLMs themselves do not have a business moat, OpenAI is focusing on the right areas to protect their business investment.

— [How OpenAI Can Secure Its Investments - A Playbook That Needs Microsoft](https://www.forbes.com/sites/lutzfinger/2023/08/18/the-potential-moat-of-openaihow-openai-can-secure-its-investments/) · _Forbes_

Question 46

What are the four approaches to addressing bias in AI models?

Accepted Answer

_Four practical methods to reduce AI bias in production systems._

First, identify the bias by testing features that shouldn't have impact, like running loan applications 100 times to see gender differences. Second, help users understand potential bias through UX, like asking specifications when someone requests an image of a nurse. Third, correct the data through prompting, prompt tuning, or fine-tuning with better examples. Fourth, create guardrails or constitutions that let AI critique itself based on rules. Each approach has limitations, and changing underlying biased datasets in large models remains hard.

— [Bias in, bias out: Addressing the impact of bias in AI](https://www.lutzfinger.com/transcript/2023-06-07-bias-in-bias-out-addressing-the-impact-of-bias-in-ai/) · _The Edge_

Question 47

What are the fundamental limits of large language models that people should understand?

Accepted Answer

_Language models face limits in reasoning, planning, and true understanding._

First, language is only part of human knowledge. We know more than we can describe. A four year old has had way more sensory impressions than any large language model, despite not being able to read. Second, LLMs can't truly reason or explain their decisions because they're built on neural networks, which work like gut feel rather than rule based systems. They predict the next word based on probability distributions, not logical reasoning. Third, there's the data quality problem. And finally, LLMs are fundamentally probabilistic, so they can't reliably predict specific individual choices or plan multi step processes independently.

— [A reality check on AI](https://www.lutzfinger.com/transcript/2024-07-10-a-reality-check-on-ai/) · _The Edge_

Question 48

What are the main barriers preventing AI from reaching AGI?

Accepted Answer

_Pattern matching alone cannot replicate human cognitive balance._

Humans can think fast and instinctively (System 1) or slowly and logically (System 2). AI only works through patterns, missing this balance. AI struggles with context and often misses important details humans pick up naturally. Current AI builds outputs on previous ones (autoregressive models), so mistakes can snowball. Many researchers believe much of this might be fixable by 2050 or 2030, but not this year. AI lacks real understanding and self-awareness.

— [No AGI But A 'Killer App' — 2025 AI Prediction](https://www.forbes.com/sites/lutzfinger/2025/01/03/no-agi-but-a-killer-app2025-ai-prediction-110/) · _Forbes_

Question 49

What are the main limitations of large language models that knowledge graphs help address?

Accepted Answer

_LLMs are powerful but flawed. Knowledge graphs provide the missing structure._

Large language models have several undesired behaviors. They hallucinate, generating false information. They need guardrails to prevent answering certain prompts inappropriately. They have gaps in their encoded knowledge and sometimes fail to retrieve information even when it exists, often due to how prompts are formulated. The prompt itself can cause the model to deprioritize relevant information. Knowledge graphs help with all these issues by adding structure and control. They make models more reliable and accurate at every step, from training to retrieval.

— [The synergy between LLMs and knowledge graphs](https://www.lutzfinger.com/transcript/2023-08-24-the-synergy-between-llms-and-knowledge-graphs/) · _The Edge_

Question 50

What are the three main components of a multimodal AI system?

Accepted Answer

_Encoders, fusion, and decoders work together to understand diverse data._

A multimodal AI system consists of three main components. First, encoders convert raw data like text, images, sound, log files, videos, IoT sensor information, and time series into vectors stored in a latent space. Second, the fusion mechanism combines multiple types of data to find what's most relevant. Third, decoders deliver information from the latent space into something we understand, for example, finding an image of a house. These components work together to process virtually any data type.

— [Multimodal AI In 2025: From Healthcare To eCommerce And Beyond](https://www.forbes.com/sites/lutzfinger/2025/01/06/multimodal-ai-in-2025-from-healthcare-to-ecommerce-and-beyond/) · _Forbes_

Question 51

What are the three main problems with training data that can cause AI to get things wrong?

Accepted Answer

_Three critical failures that undermine machine learning accuracy and fairness._

Training data can fail in three ways. First, it might not be sufficient. The data may lack context or situations the algorithm will encounter in the real world. Second, it might be biased. This means data is systematically missing certain perspectives or groups. For example, if texts predominantly show nurses as women, the algorithm learns that pattern without understanding societal goals like gender equality. Third, the data might be outright wrong. Labels may be inaccurate or based on performances rather than reality, leading algorithms to learn incorrect associations.

— [It's The Data, Stupid! Why AI Might Get It Wrong.](https://www.forbes.com/sites/lutzfinger/2020/05/28/its-the-data-stupid-why-ai-might-get-it-wrong/) · _Forbes_

Question 52

What are the two main challenges value-based care aggregators must overcome?

Accepted Answer

_Small patient pools and personalization create unique VBC obstacles._

Value-based care aggregators face two key challenges. First is the issue of small numbers. Self-insured companies can't offer a large patient pool, only their own employees. VBC providers willing to risk fees with 66 million Medicare patients may see their approaches work on average, but with only 100 employees, those averages might not hold. Second is personalizing healthcare. While treatment is standardized, behavioral changes are unique to each individual. Weight loss alone has over 77,000 books, showing no one-size-fits-all solution exists.

— [Changing Healthcare - A Platform Approach Using Data & AI](https://www.forbes.com/sites/lutzfinger/2023/07/05/changing-healthcarea-platform-approach-using-data--ai/) · _Forbes_

Question 53

What did you get wrong about multimodal AI interfaces?

Accepted Answer

_Right about multimodal potential, but overly optimistic on timing and control._

We were right that models can encode anything into a latent representation, like Esperanto for computers, and switch back and forth between text, voice, and images. We saw this with clip, voice to voice, encoding from one language into voice representation of another. But we missed the timeline. It's not as strong as we hoped. The challenge is LLMs are only probability distributions. They're not controllable. If you have a machine that's not controllable, it sucks as a machine. That's okay for your friend at the bar at 2 a.m., not your computer.

— [Did our predictions hold true?](https://www.lutzfinger.com/transcript/2024-03-27-did-our-predictions-hold-true/) · _The Edge_

Question 54

What do companies need to win the AI search war?

Accepted Answer

_User engagement and understanding context are keys to winning search._

Like in the 1990s, it all boils down to users. If a search tool can grab enough attention, keep people engaged, and build loyalty, it has a shot at success. Companies need to grasp the concept of the long tail in search. Human queries are inherently complex, which is why keyword-based search initially struggled. Understanding context while providing the right level of information is a challenging task that winners must master.

— [2025: The Year Of Search But Not The End Of Google](https://www.forbes.com/sites/lutzfinger/2025/01/03/2025-the-year-of-search-but-not-the-end-of-google/) · _Forbes_

Question 55

What do you mean when you say winners won't be those who use AI, but those who redesign work around it?

Accepted Answer

_Business needs remain constant, workflows must evolve._

Our business hasn't changed. We still have needs, we still travel, we still consume, we still build. The only question is how can I do this business more effectively? Yes, there will be people who use ChatGPT to write an email, but the winners will be those who either build the platforms or rearrange the workflows in order to be effective. The focus is on the workflow, the surrounding, the setup. How do we empower you in your environment, not can you use it?

— [From Google to Cornell: Rethinking AI Workflows | Lutz Finger | Build AI Podcast 17](https://www.lutzfinger.com/transcript/2026-04-16-from-google-to-cornell-rethinking-ai-workflows-lutz-finger-build-ai-podcast-17/) · _Build AI_

Question 56

What does 'bring your own data' mean for competitive advantage with LLMs?

Accepted Answer

_Proprietary data access separates sustainable advantages from temporary ones._

In this new world of foundation models, data access is key. Bring your own data will be a game-changer for this young industry. Bloomberg launched a massive financial LLM only they could have trained because they had proprietary data access. Data moats offer a more sustainable advantage, and exclusive access to proprietary data can be a significant barrier to entry. But data in itself is not helpful, data needs to create value. This contrasts with scraped internet data that anyone can access.

— [What Is The Competitive Advantage Of LLMs Like ChatGPT For Your Business? Three Takeaways.](https://www.forbes.com/sites/lutzfinger/2023/04/04/what-is-the-competitive-advantage-of-llms-like-chatgpt-for-your-business-three-takeaways/) · _Forbes_

Question 57

What does it mean for a company to be AI-native versus just using AI?

Accepted Answer

_AI-native means AI is the core product, not just an enhancement_

An AI-native company means the primary product offering or value is driven by AI, not just enhanced by it. Netflix is a channel for videos that uses AI to suggest movies, but you didn't need AI for Netflix. LinkedIn is a social network that uses AI to connect people, but you didn't need AI for LinkedIn. These are not AI-native. Companies like 11 Labs, Midjourney, or Cursor are AI-native because AI is at the heart of what's being built, what's being sold, and the people you hire early on.

— [The Future of AI, with Martin Casado](https://www.lutzfinger.com/transcript/2025-03-20-the-future-of-ai-with-martin-casado/) · _The Keynote on AI_

Question 58

What does Sam Altman mean when he says 'we know how to build AGI'?

Accepted Answer

_Altman likely refers to OpenAI's o1 model and iterative reasoning._

When Altman says we know how to build AGI, he is probably referring to OpenAI's o1, a model designed for reasoning through an iterative, self-calling process. The AI has two dedicated steps: Iteration and Reflection, where the model generates an output, evaluates it, and refines it, and a Feedback Loop, where the model revisits outputs, critiques them, and improves them further. This represents a paradigm shift. While OpenAI had previously focused on bigger models, they are now focused on thinking longer by doing this type of inference loop.

— [Sam Altman Says, 'We Know How To Build AGI' — Not So Fast](https://www.forbes.com/sites/lutzfinger/2025/01/07/sam-altman---agi--ai-workforce-in-2025-the-battle-of-tech-giants/) · _Forbes_

Question 59

What does the article say we need to address AI data problems going forward?

Accepted Answer

_No quick solutions exist; fairness itself remains hard to define precisely._

There's no easy fix coming. Beyond technical difficulties, researchers struggle with defining what is fair. We need tools to check for biases and missing data in our models. We also need the right incentives to check for drift and update algorithms as society changes. Drift happens when datasets become biased over time even if initially unbiased, because society and beliefs evolve. Machine learning and AI will be powerful forces for good, but only if the training data is correct and fair.

— [It's The Data, Stupid! Why AI Might Get It Wrong.](https://www.forbes.com/sites/lutzfinger/2020/05/28/its-the-data-stupid-why-ai-might-get-it-wrong/) · _Forbes_

Question 60

What does the term agents actually mean in AI, and is the technology working?

Accepted Answer

_Agents has three definitions but the technology isn't quite there yet_

Agents is one of the most diluted terms, almost useless. There are three working definitions. First, a virtual agent representation that helps with support tickets. Second, the technical definition where you close the control loop with AI so output becomes input, doing multiple iterations without humans to solve higher level tasks, like booking a ticket. Those aren't really working yet. The tech isn't quite there, and these elements are very unpredictable for complex tasks. Third, any time an LLM calls out to a tool, like sending email. The term itself is almost useless, but the things it's talking about are very real trends.

— [The Future of AI, with Martin Casado](https://www.lutzfinger.com/transcript/2025-03-20-the-future-of-ai-with-martin-casado/) · _The Keynote on AI_

Question 61

What does true swarming mean, and how is it different from choreography?

Accepted Answer

_Real swarms have no leader; each drone decides independently._

Some companies have drones flying in formation or following a leader, self-healing formations that rearrange when members are shut down. I consider that choreography, not actual swarming. True swarming means every single member of the group knows everything about all the other members and makes its own independent decisions. There is no leader, no single point of failure. They all know the objectives and negotiate among themselves who does what to achieve it. It is a swarm where they are all equal partners, just with different capabilities.

— [AI on the Front Lines, with Alexander Fink](https://www.lutzfinger.com/transcript/2026-04-30-ai-on-the-front-lines-with-alexander-fink/) · _The Keynote on AI_

Question 62

What framework does Lutz use to analyze AI products?

Accepted Answer

_Three dimensions reveal how AI products really work._

I analyze any AI product by three factors: control, data, and transparency. Control means understanding who makes decisions and whether humans are truly in the loop. Data looks at who owns and has access to the information that powers AI. Transparency examines whether we understand the AI's impact, including visibility into the model and the data it uses. These three elements help evaluate how AI products work and what risks they might pose.

— [Biden's AI Plan And Why Central Control Might Not Work](https://www.forbes.com/sites/lutzfinger/2024/10/26/bidens-ai-plan--why-central-control-might-not-work/) · _Forbes_

Question 63

What framework helps assess whether imperfect AI decisions are acceptable?

Accepted Answer

_Match your AI quality threshold to the stakes of the decision._

The framework asks: How good is your decision quality, and how important is the decision? How bad is a false decision? We discussed two opposite examples at Station F. One company did predictive maintenance for infrastructure where errors are costly, requiring high quality predictions. Another did social media outreach for hiring where low quality matters less since you already have low close rates. You can mitigate both. For low-stakes, automate with feedback loops. For high-stakes infrastructure, use augmentation first, guiding engineers and creating Jira tickets.

— [Building AI products: 5 lessons from our founders' workshop](https://www.lutzfinger.com/transcript/2024-08-01-building-ai-products-5-lessons-from-our-founders-workshop/) · _The Edge_

Question 64

What is a data product?

Accepted Answer

_Definition and how to build one._

> Now, to be a good product developer, you need to think about self-referential loops within your products. How to get signal, how to get feedback from your users so your AI, or your data science, can leverage that information to make the product work better. So that moved a little bit within it. The innovation, and the inspiration about, and the curiosity about how to use data in novel manners, that's still data science, because that's…

— [Unit 3 Full Transcript](https://www.lutzfinger.com/ecornell/) · _eCornell course_

Q&A, AI Product

Can a well-designed prompt outperform OpenAI's o1 model?

Does Google still have a chance against the new AI search competitors?

How accurate are current deepfake detection technologies?

How are AI agents different from RPA (robotic process automation)?

How can AI support frontline workers who aren't sitting at computers all day?

How can startups use large language models to improve their products?

How can transparency tools like NFTs help combat deepfakes?

How did the Apple Card algorithm discriminate despite not using gender as a direct feature?

How did you replace yourself with a bot in your AI and product course?

How do chained AI models create better workflows than simple generative AI tools?

How do encoder-decoder pairs work to create deepfakes?

How do Generative Adversarial Networks make deepfakes more realistic?

How do you decide if AI is the right fit for your product problem?

How does AI change healthcare?

How does AI change the way we interact with computers compared to previous interfaces?

How does AI enable better information retrieval than traditional search?

How does AI find patterns in data that humans miss, like in healthcare?

How does Catalog ensure certain workflows always follow the same steps when using LLMs?

How does Catalog prevent large language models from hallucinating when retrieving enterprise information?

How does Catalog use large language models to simplify enterprise software access?

How does jamming affect drone operations, and why does it make autonomy necessary?

How does LegalOS use LLMs to make legal knowledge more accessible?

How does Marpai Connect use AI to overcome the small numbers problem?

How does multimodal AI improve the healthcare experience for doctors dealing with overwhelming patient data?

How does Neuralens measure brain health non-invasively compared to current methods?

How does Think Labs AI use physics informed AI differently than general purpose models like ChatGPT?

How does TikTok's algorithm differ from Clubhouse's approach to content discovery?

How does your company Gooder AI address the gap between technical metrics and business value?

How has natural language processing changed since 2015, and why does that matter for legal tech?

How has online shopping changed over the past 25 years?

How has the shift from SEO to GEO changed what websites need to do for brands?

How is OpenAI trying to improve user experience to build stickiness?

How is r2decide using multimodality to transform the online shopping experience?

How should product builders think about data bias in AI systems?

How will Large Language Models impact healthcare according to Lutz?

If agents are the future, why won't one giant model just do everything?

If OpenAI doesn't have a model moat, what kind of moat does it actually have?

Is ChatGPT actually at risk of shutting down despite the bankruptcy headlines?

Isn't OpenAI's o1 model a first step toward AGI?

Should enterprises invest equally in predictive and generative AI, or focus on one?

Should founders focus on building new AI models or developing applications with existing technology?

Should you trust OpenAI blindly?

What advantage do existing platforms like Google, Meta, and TikTok have in the bot space?

What are AI-generated nudges and how do they work in r2decide's system?

What are the five strategies OpenAI is using to create a business moat?

What are the four approaches to addressing bias in AI models?

What are the fundamental limits of large language models that people should understand?

What are the main barriers preventing AI from reaching AGI?

What are the main limitations of large language models that knowledge graphs help address?

What are the three main components of a multimodal AI system?

What are the three main problems with training data that can cause AI to get things wrong?

What are the two main challenges value-based care aggregators must overcome?

What did you get wrong about multimodal AI interfaces?

What do companies need to win the AI search war?

What do you mean when you say winners won't be those who use AI, but those who redesign work around it?

What does 'bring your own data' mean for competitive advantage with LLMs?

What does it mean for a company to be AI-native versus just using AI?

What does Sam Altman mean when he says 'we know how to build AGI'?

What does the article say we need to address AI data problems going forward?

What does the term agents actually mean in AI, and is the technology working?

What does true swarming mean, and how is it different from choreography?

What framework does Lutz use to analyze AI products?

What framework helps assess whether imperfect AI decisions are acceptable?

What is a data product?

What is a knowledge graph and how does it differ from a taxonomy?

What is an AI agent, really?

What is LIME and how does it help with explainable AI?

What is the augmentation thesis for AI in workflows?

What is the BizML framework and why is it needed for AI projects?

What is the core discovery mechanism behind multimodal AI, and how does it relate to Amazon's recommendation approach?

What is the difference between LLMs and traditional machine learning models in healthcare?

What is the difference between using a deepfake model and training one?

What is the first-order motion model approach to creating deepfakes?

What is the hallucination problem with large language models and why does it matter for businesses?

What is the killer app of AI?

What is the minimum viable quality concept when building AI products?

What is the missing moat in AI?

What legitimate uses do deepfakes have in the entertainment industry?

What level of autonomy does Swarmar allow in target engagement, and where do humans decide?