Welcome, executives and professionals.

We identify and breakdown the top 1% of Generative AI for enterprises.

This week:

• Capgemini explores GenAI through industry leader perspectives.

• Deloitte AI Institute on how AI agents are reshaping work.

• How Airbnb evolved its Automation Platform for GenAI.

• Creating 2.7 million property description summaries with GenAI.

• Fast Fives: Transformation and technology in the news this week.

• Career opportunities & events.

Read time: 4 minutes.

BEST PRACTICE INSIGHT

Capgemini explores GenAI through the perspectives of industry leaders

Brief: Capgemini published a new 150-page report, ‘Gener(AI)ting the Future’ which explores the potential of GenAI to impact enterprises and society, through various influential perspectives.

Breakdown:

• The report features insights from pioneering founders and leaders in business, policy, and academia. For instance, Mistral AI CEO Arthur Mensch: “The rare talent that we recommend every organization look for is the software engineer who can also do data science.”

• AI pioneer Andrew Ng: “For many jobs, AI will only automate or augment 20-30% of tasks. So, there's a huge productivity boost, but people are still required for the remaining 70% of the role.”

• Capgemini CEO Aiman Ezzat: “Any use case that requires high performance or deep domain expertise will likely continue to go down the path of specialized models.“

• Over the past year, GenAI adoption has increased across various domains (sales and marketing, IT, operations, R&D, finance, and logistics), reshaping roles for employees to concentrate on more complex strategic tasks.

• Early adopters of GenAI are reporting positive outcomes, such as improved operational efficiency and enhanced customer experiences, though ethical considerations remain critical to prevent misuse.

• The report highlights GenAI's role in tackling climate change, emphasizing the advantages of small language models (SLMs) and includes recommendations for successful GenAI implementation.

Why it’s important: Capgemini's report highlights the transformative potential of GenAI across industries and the importance of strategic leadership and ethical practices to maximize its benefits.

BEST PRACTICE INSIGHT

Deloitte AI Institute on how AI agents are reshaping work

Brief: Deloitte AI Institute’s new 18-page report ‘Prompting for Action’ explores how AI agents expand capabilities, use cases, and enterprise impact from Generative AI.

Breakdown:

• The report compares AI agents to typical language models, highlighting their differentiated ability to not only interact but also reason and act more effectively. It examines scope, planning, memory, tools, data, and accuracy.

• It discusses how multi-agent systems of role-specific AI agents coordinate and collaborate with humans to orchestrate complex work, highlighting their key benefits for enterprises, such as accuracy enabled by “validator” agents.

• A real-world example visually breaks down how multi-agent systems enhance speed, efficiency, and scalability compared to traditional research projects.

• The report also explores four key use cases for enterprises today: individualized financial advisory, dynamic pricing and personalized promotions, talent acquisition, and personalized customer support.

• It outlines implications for strategy, risk, talent, technology, and data, including emerging risks like “agent autonomy” and potential unintended consequences from minimal human oversight.

Why it’s important:

Leaders are recommended to prioritize use cases and develop a strategic roadmap for AI agents, while addressing infrastructure, talent, data governance, and culture.

CASE STUDY

How Airbnb evolved its Automation Platform for GenAI

Brief: Airbnb's case study highlights the evolution of its Automation Platform, moving from Version 1 with static workflows for conversational systems to Version 2, which supports large language model (LLM) applications.

Breakdown:

• The initial platform version supported traditional conversational AI products but faced challenges including limited flexibility and scalability issues.

• Experiments showed that LLM-powered conversations provide a more natural and intelligent user experience than more rules-based workflows, enabling open-ended dialogues and better understanding of nuanced queries.

• Despite benefits, LLM applications are still evolving for production, such as reducing latency and minimizing hallucinations. These limitations affect their suitability for some large-scale, high-stakes scenarios involving millions of Airbnb customers.

• For sensitive processes like claims processing that need strict data validation, traditional workflows are considered more reliable than LLMs.

• Airbnb combines LLMs with traditional workflows to leverage the strengths of both approaches and enhance overall performance.

• The upgraded platform includes capabilities to facilitate LLM application development, featuring capabilities such as chain of thought, context management, guardrails and observability.

Why it’s important: Airbnb's Automation Platform evolution demonstrates the benefits of merging more traditional rules-based workflows with LLM technology to improve user experience and operational efficiency. Detailed case study summary here.

CASE STUDY

How Booking.com creates 2.7 million property description summaries with GenAI

Brief: AWS released a 40-minute video on taking GenAI from idea to production. 28:57 onwards details how Booking(dot)com utilizes GenAI to produce 2.7 million property description summaries, improving user experience for over 100 million mobile app users.

Breakdown:

• The primary objective is to enable users to quickly read property descriptions at a glance on mobile devices. Each property description, typically around five paragraphs, is summarized down to 235 words.

• Booking(dot)com’s fine-tuning strategy starts with a general summary (Step 1), to a segment-based personalization summary using traveler type (Step 2), then a search-based personalization summary utilizing user preferences (Step 3).

• Step 1 summaries are produced by leveraging human-in-the-loop data correction, fine-tuning LLMs, evaluating models, applying direct preference optimization (DPO), deploying the model, and conducting A/B testing.

• Fine-tuning improved quality by 16%, reduced costs by 6x, and enhanced throughput by 8x.

• Future plans include automatic updates, implementing ‘Step 2’ personalized fine-tuning and deploying real-time inference.

Why it’s important: The ability to generate concise and personalized property summaries through GenAI enhances user experience. These innovations help position booking(dot)com to stay competitive in the travel industry.

Transformation

In the news this week:

1. McKinsey published an article on AI power demand, highlighting opportunities for enterprises and investors across the value chain. A separate blog examines the full supply chain involved in producing today’s GenAI.

2. Deutsche Bank released a 43-page report on adopting GenAI in banking. It includes building a use case portfolio and insights on the importance of data quality, business domain expertise, human-in-the-loop processes, and upskilling.

3. Accenture discusses agentic architecture and implementation. A different blog provides a breakdown of AI agents, from basics to multi-agent systems, with code snippets for practical application.

4. Stanford's Cyber Policy Center released a 470-page report analyzing existing proposals for governmental policy and regulation of GenAI.

5. Bain explores how China’s strength in e-commerce and research, along with high consumer trust in AI, suggests GenAI could expand rapidly in Chinese retail.

Technology

In the news this week:

1. Google launched 'Grounding with Google Search' for its Gemini API and AI Studio, allowing developers to add real-time search results to model responses, reducing hallucinations and improving accuracy. CEO Sundar Pichai also shared that AI now writes over 25% of Google’s code, with human oversight.

2. Microsoft's GitHub will add Anthropic and Google models to its AI coding assistant Copilot.

3. OpenAI launched ChatGPT Search and is also reportedly developing its first custom AI chip. CFO Sarah Friar revealed a 5-6% conversion rate from free users to paying ChatGPT subscribers.

4. Anthropic has released PDF support for its Claude 3.5 Sonnet model in public beta, enabling analysis of both text and visuals, including charts and images, within large documents.

5. Meta's open-source Llama model has reportedly been used by Chinese military researchers to create ChatBIT, an AI tool for intelligence analysis and strategic planning.

Career opportunities

• JPMorgan - Executive Director, Applied AI/ML

• AWS - Principal Solutions Architect, Generative AI Startups

• AIG - Vice President, SWE Generative AI

Events

• Columbia University with Accenture - Designing Generative AI Platforms for Enterprises (1 hour) - November 12, 2024

• NVIDIA - Enhance Generative AI Model Accuracy (1 hour, Virtual) - November 14

Whenever you’re ready, here’s how we can help you:

1. Get 15 proven Generative AI case studies from industry leading enterprises for free. In-depth analysis and tangible value delivered.

Or send feedback directly, we’d love to hear from you. Your insights help us enhance your experience.

All the best for the week ahead,

Found this valuable? Share with a colleague.
Received this email from someone else? Sign up here
Let's connect on LinkedIn