MLOps Live

Building Scalable Customer-Facing Gen AI Applications Effectively & Responsibly with MongoDB - July 30th, 2024

Transforming Enterprise Operations with Gen AI

Alexandra Quinn | June 24, 2024

Enterprises are beginning to implement gen AI across use cases, realizing its enormous potential to deliver value. Since we are all charting new technological waters, being mindful of recommended strategies, pitfalls to avoid and lessons learned can assist with the process and help drive business impact and productivity. In this blog post, we provide a number of frameworks that can help enterprises effectively implement and scale gen AI while avoiding risk. We also include a number of use cases, from R&D to automotive to the supply chain. In the end, we list potential hurdles and how to overcome them.

This blog post is based on the webinar “Transforming Enterprise Operations with Gen AI”, which was held with Dinu de Kroon, Partner and Operations Hub Lead, Nicola Unfer, Sr. Program Delivery Analyst and Davide Di Lucca, Research Science Analyst, from McKinsey, and Yaron Haviv, Co-Founder and CTO, Iguazio (acquired by McKinsey). You can watch the webinar recording here.

5 Questions to Ask and Answer About Gen AI in the Enterprise

The evolution of AI began in the 1950s, but the advent of ChatGPT and other generative AI capabilities have created the “perfect storm” of AI. This revolution has been driven by the convergence of massive computing power, enabling data processing at unprecedented scale and speed; an abundance of data available through the internet for model training; and pre-trained transformers that empower us to efficiently work with unstructured data.

At this point in time, it’s recommended that enterprises ask themselves five questions about operationalizing gen AI:

  1. Where do we stand on the journey to generate value from gen AI? 
  2. Are we ambitious enough with generative AI?
  3. Where does the money sit in operations?
  4. What does it take to scale AI/gen AI in operations?
  5. How to stop using risk as an excuse?

Let’s dive into each one.

1. Where do we stand on the journey to generate value from gen AI?

Both Gen AI and AI provide significant value. Each has its own strengths and integrating them also brings value.

Gen AI“Traditional” AI
Generates new data by learning from large data sets and identifies patterns within. Capabilities include text, visual, sound, etc.Output can be unstructured dataOpen-ended and creative, “human-like” behaviorSolves specific tasks by making predictions based on previously analyzed sets of data and predefined rulesOnly structured data as outputGoal-oriented and specific
Use cases:Design conceptsCode generationMarketing or social media copyUse cases:Forecasting salesSupply chain optimizationPredictive maintenance

2. Are we ambitious enough with generative AI?

According to McKinsey & Company, only 1 out of 10 organizations focuses on business model reimagination. But the gen AI evolution is just at the beginning and the art of possible has never been bigger. Organizations are beginning to tap into the value of gen AI, either capturing low hanging fruit or realizing its full-blown potential.

Gen AI innovation is mostly present in consumer-facing industries. For example:

  • Procurement: Automating supplier negotiations using internal requirements and external data
  • Manufacturing: Using synthetic data to retrain vision system models and optimize workflow of the conveyor system
  • Supply Chain: Answering complex supply chain questions from employees on messaging tools like Microsoft Teams, Slack, or Whatsapp
  • Consumer Marketing: Producing videos to answer common customer questions, interact with customers and even help them build their shopping lists

Overall, gen AI is expected to have a significant productivity impact across all industry sectors, from high-tech to retail to banking to energy to agriculture, and many more. In addition, entire industries will be reimagined and reshaped.

Source: McKinsey report: The Economic Potential of Generative AI: The Next Productivity Frontier

Within industries and organizations, gen AI value comes mainly from operations, marketing and sales, and engineering. Each role type constitutes approximately one third of the value.

It’s also important to remember that generative AI is only one piece of the pie and should be combined with additional technologies, like AI, AR/VR and Web 3.0. According to McKinsey, the value potential of combining gen AI with traditional AI is $17-26 trillion, with gen AI constituting 20-40% of that value.

3. Where does the money sit in operations?

Gen AI offers value across entire domains and functions.

Spotlight1: R&D Use Case

Gen AI can help optimize, automate and innovate across multiple R&D steps, including:

  • Automating steps in R&D development (e.g., new chemical compositions, circuit designs)
  • Optimizing traditional part designs (e.g. component weight reduction)
  • Accelerating coding and overall software generation 
  • Automating the conversion of code from one language to another 
  • Accelerating R&D process through R&D Virtual Expert leveraging internal and external insights 
  • Improving product requirements from customer claims and regulatory updates

Spotlight2: Automotive ROI

Based on an ROI analysis, McKinsey found that for a leading automotive OEM, gen AI is expected to lead to an estimated long-term efficiency potential of 21-25% in indirect functions, within the next few years. This includes research and development, indirect production and procurement. Across these functions, gen AI will affect 70-80% of all activities.

See more use cases and productivity benefits in the webinar.

Best 10 Free Datasets for Manufacturing [UPDATED]

Let's Discuss Your Gen AI for Manufacturing Use Case

Streamline the way your enterprise builds, operationalizes and scales generative AI applications.

4. What does it take to scale AI/gen AI in operations?

While use case identification is important, organizations should also extend their thinking to the long-term, strategic gen AU vision.

Ask yourself:

  • Do I consider gen AI an opportunity or a threat?
  • What use cases do I have? Where do I start - operations, customer service, supply chain, something else?
  • Is my organization ready? What are the tech considerations? What talent do I require?
  • What is the long-term vision? How can I leverage this across my organization? How will this impact my operating model?

Answering these questions will increase the likelihood of gen AI success.

5. How to stop using risk as an excuse?

Responsible AI is the framework for ensuring AI is developed and deployed in a safe, trustworthy and ethical manner. As of now, we only partially understand how LLMs work. This means there are risks that need to be taken into consideration and handled.

Responsible AI consists of establishing policies, best practices and tools to ensure: 

  • Human-centric AI development and deployment that embeds human oversight, includes diverse perspectives and aligns with organizational values 
  • Fair, trustworthy and inclusive AI to prevent bias and discrimination 
  • Transparent and explainable AI enabling impacted individuals (e.g., developers, users) to understand how systems work and how decisions are made 
  • Robust data protection, privacy and security measures to protect sensitive information (e.g., PII) 
  • Ongoing monitoring and evaluation of AI systems to ensure they continuously meet ethical, legal and social standards

3 Gen AI Supply Chain Use Cases

Let’s go from theory to practice and see how gen AI can provide value across three use cases, relating to the supply chain and procurement verticals.

1. Augmented Operator

This use case covers a manufacturing process with multiple machines. Tackling why machines breakdown is a complex process. It requires investigating multiple parameters, checking the machine’s manuals and assistant tools, calling the supervisor, assessing supplies, etc. 

A gen AI-powered maintenance advisor can:

  • Understand the capabilities of the machine’s digital assisting manual and tool 
  • Explain the type of issue that caused the machine to break
  • Suggest the most probable root cause of the defect
  • Provide actionable troubleshooting suggestions
  • Support the operator carrying out maintenance activities
  • And more

2. ContractAI

The second use case helps analyze procurement contracts. Contract analysis is a manual, time-consuming and expensive activity that involves assessing the clauses of the contract and identifying loopholes and discrepancies.

ContractAI can assist with:

  • Ranking contracts based on criteria like expiration date and performance
  • Creating contract summaries
  • Highlighting specific sections like terms and conditions
  • Providing suggestions for improving contracts
  • And more

3. Control Tower Co-pilot

In the supply chain landscape, companies struggle to deal with supply chain volatility, and especially disruptions and inflation. A co-pilot can help solve complex tasks based on KPIs and inputted data. The co-pilot can provide analysis information on:

  • Lowest and highest performing regions
  • Reasons for low production, like non-optimal inventory allocation
  • How to track issues and solutions
  • Expected impact of potential scenarios
  • And more

The 4 Transformation and Implementation Hurdles

To drive impact along the phases of the gen AI journey, organizations need to overcome four hurdles:

  1. Failure to scale at the network level
  2. Failure to pilot with effective testing and transformation initiation
  3. Failure to start and launch the transformation
  4. Failure to plan and adequately design the transformation

6 Core Gen AI Transformation Enablers

To overcome these hurdles and get started, the following framework can help “rewire” the organization.

  1. Strategic roadmap - Aligning the gen AI strategy with technology aspirations, while ensuring the transformation captures value and unlocks a competitive advantage.
  2. Talent - Managing talent and staying ahead of the gen AI skill gaps.
  3. Op Model - Organizing the org and teams to deliver on the gen AI strategy.
  4. Technology - Setting up a scalable tech stack and infrastructure to support multiple gen AI use cases and solutions.
  5. Data - Setting up a robust data foundation to scale gen AI across the organization.
  6. Adoption and Scaling - Designing the scaling plan to ensure easy re-usability and scalability across the organization, delivering effective training to support skill building, managing culture change at scale and addressing risk and responsible use of gen AI across the organization.

Tips for Getting Started

Based on our experience in the field, here are six tips to get started with delivering gen AI impact:

  • Start with bold business opportunities and strategy setting - Set a bold aspiration for the business and work backward to potential AI and gen AI applications. Identify your strategy. Avoid tech for tech's sake.
  • It is both the “what” and the “how” - Deploying gen AI across select priority domains is just the first step, architecting a gen AI stack that is robust, cost efficient and scalable for years to come is key.
  • Speed is a strategy as companies are “learning how to learn” - Improve your learning quotient, evolve with the industry and avoid one-way doors,
  • The partnership landscape is complex and shifting - The ecosystem continues to expand with leadership often shifting; big tech/FM players now recognize the importance of an end-to-end solution for their customers to capture value.
  • Be ready to redesign processes and manage change - Models and technology are required but not sufficient – process redesign and change management is key to adoption. Every $1 in tech requires another $5 in change.
  • Play offense when it comes to typical challenges to capturing value - Realizing P&L impact from gen AI requires close partnership with HR, finance, legal, and risk to change the resourcing strategy and productivity expectations of the organization

Next Steps

Transforming your enterprise operations with gen AI and bringing business value requires strategic planning -- from technological aspirations to talent to resources to scale. The frameworks and use cases above are key to ensuring long-term success and impact across the enterprise. Iguazio’s gen AI factory can help operationalize and derisk gen AI and AI at scale across the enterprise, effectively and responsibly. For more information on how to get started, click here.