Generative Artificial Intelligence in Business
0 likes664 views
My invited lecture at the Indian School of Business (ISB) for the participants of the Emerging Leaders Program (ELP), Batch 8 on Jan 5, 2025. In this talk, I discuss the key issues around adoption of Generative AI in Business, and additionally bring the learnings from my research on what are the impediments, the failure modes and the measures of value in business.
![[DSC DACH 23] ChatGPT and Beyond: How generative AI is Changing the way peopl...](https://cdn.slidesharecdn.com/ss_thumbnails/chatgptandbeyond-230424074311-fa7c1960-thumbnail.jpg?width=560&fit=bounds)
Ad
More Related Content
What's hot (20)
Similar to Generative Artificial Intelligence in Business (20)
Ad
More from Dr. Tathagat Varma (20)
Ad
Recently uploaded (20)
Generative Artificial Intelligence in Business
- 1. GenAI in Business DR. TATHAGAT VARMA โCOGNITIVE CHASMS: A GROUNDED THEORY OF GENAI ADOPTIONโ
- 3. GenAI: The New Shiny Penny? ๏ต Several Trillion $$$s of economic impact projected. ๏ต 30% growth expected in 2024: Over $20bn VC funding in the first three quarters. ๏ต Capex of $1tn+ expected in the coming years!
- 4. GenAI Hype Source: Gartner, https://www.gartner.com/en/articles/hype-cycle-for-genai
- 5. What is Generative AI? Ability to create โnewโ content using simple prompts in English Content could include text, images, audio and video, etc. Utilizes advanced neural network architectures Leverages vast amount of existing data for pre-training
- 6. Types of Generative AI Generative Adversarial Networks (GANs): Two neural networks, the generator and the discriminator, compete against each other to create new outputs. Variational Autoencoders (VAEs): Good for exploring variations in data in a specific direction. Diffusion models: Create detailed and complex images by gradually refining random noise into structured images. Transformer models: Use a self- attention mechanism and a feed-forward network to identify and memorize patterns in data. Deep Generative Models (DGMs): A combination of a generative model and a neural network. Autoregressive models: Suitable for tasks such as language generation and image synthesis. Large Language Models (LLMs): Statistical models of language that estimate the probability of a sequence of words appearing within a particular context.
- 7. Popular Toolsโฆgrowing daily! Text โข Conversation: ChatGPT, Claude โข Content: Jasper โข Writing: Grammarly, Anyword โข Email: Shortwave โข Notes: Mem Image โข Generation: Dall-E, Midjourney โข Enhancement: Adobe Photoshop Audio โข Voice Generation: Wellsaid, Respeecher, Altered, Murf โข Voice Cloning: ElevenLabs โข Music: Suno โข Text-to- Speech: Speechify, Natural Reader Video โข Generation: Runway โข Editing: Descript โข Social Media: Flicki โข Digital Avatars: Synthesia โข Content Marketing: Peech โข Script to video: Visla Technology โข Coding and debugging: Copilot โข App building: Microsoft Power Apps, Wix, Framer, Pico
- 8. Key Benefits Productivity Automate tasks that are manual, boring, repetitive, dangerous Allow humans to focus on higher value-adding tasks Efficiency Reduce lead time, process overheads, wastage Optimize operational costs and improve bottomline Experiences Deliver personalized experiences Enhance creativity Innovation Design new products, services and experiences Improve topline and innovate the business model Decision-making Enhance decision- making based on data-driven insights Ability to incorporate pervasive emergent learning and achieve market-scale in near- real-time
- 9. Major Limitations Bias Accuracy Ethical Cost Data
- 11. Industriesโฆvirtually every industry! Healthcare: Diagnostics, Patient Care, Retail: Inventory management, Pricing, Virtual try- ons, etc. Entertainment: Scriptwriting, Animation, Office: Productivity solutions
- 12. Functionsโฆliterally every function! Customer Service Chatbots Marketing Content creation: Ads, Blogs, Creatives Personalized campaigns, Data Analysis, Culling insights, etc. Product Development Prototyping, Design Technology: automated coding and debugging People AI-driven individual learning journeys Hiring, career progression, etc.
- 13. GenAI Value Spectrum Optimize (โ10%โ) โขTask-level adoption and process-level โvertical scalingโ lead to individual productivity hacks that might make no real impact for the firm, while process efficiency could improve the bottomline. Innovate (โ1X+โ) โขFirm-level โhorizontal scalingโ across the entire value chain leads to business value realization and superior customer experiences that improve the topline better than the normal growth. Transform (โ10Xโ) โขHarness the power of data- informed algorithmically- supported โcognitive decision- makingโ to transform the business model and deliver exceptional human experiences that are order of magnitude superior. Evolve (โโโ) โขIndustrial and societal adoption lead to market disruptions and planet-scale solutions for social good when applied judiciously with appropriate human controls, but might wreak havoc when applied indiscriminately without adequate forethought and safeguards.
- 14. Theory of Cognitive Chasmsยฉ
- 15. Theory - of - Cognitive Chasmsยฉ Coming in 2025โฆ Follow updates at: https://www.linkedin.com/company/cognitivechasm
- 16. Recap Enterprise-wide Adoption: GenAI is a โgeneral purpose technologyโ. To realize its full value, it must be seen as a โhorizontal technologyโ, i.e., enterprise-wide adoption. Strategic Value Creation: Successful adoption must strategically integrate GenAI to a firmโs value chain and deliver tangible value to the firm. True value of GenAI lies in transforming the internal capabilities and business model of the firm. Transformation Roadmap: Theory of Cognitive Chasms provides a roadmap for firms looking at systematic change management while mitigating risks and delivering tangible value.