State of AI 2025 Mckinsey: Brief summary

State of AI 2025 Mckinsey: Brief summary

Recently read the paper “State of AI 2025 by McKinsey. Very detailed analysis of the responses from many In this Blog, I want to highlight few points of high relevance as you advise and implement AI solutions to your company or customers.

1. One of the major highlights from the report is the steady increase in the use of Gen AI across enterprises since last year. This is a good sign, as predicted by our @aiThoughts.Org think tank.

2. Marketing and sales, Service Operations, and Software Development are the top functions using GenAI. We need to go a few levels deeper and peel the onion to see what kind of applications in sales and marketing are being used. For e.g., how much is the domain content in these applications? For example, Investment Banking, Supply Chain, Pharma, etc. There was mention of the use of GenAI to generate images for marketing brochures, etc, which makes me assume less domain content.

3. The issue of hallucination/inaccuracy still seems to be a major concern. All the respondents say they are manually validating at least some % of the GenAI-generated output before it is sent to the end customer. Almost two-thirds check more than 50% of the output, and one-third check 100% of the content. Many AI Gurus are saying that LLM-based models will continue to have this issue even in newer versions with large scaling. This remains the major concern for the enterprises.

4. Risk and Governance seem to be at least considered. The top 3 risks identified are Inaccuracy, IP infringement, and privacy. Unfortunately, only very large enterprises are putting a formal plan and resources to do risk mitigation. This area is a major concern.

5. While there are positive reports on cost savings at a sub-function level, almost all respondents felt that overall Enterprise level EBIT has not increased due to GenAI initiatives. There is a huge expectation that the step from Cloud, Mobile, IOT ( Digital) transformation to AI Transformation and very quickly the AI community need to start proving this giant step. Otherwise, the trend of positive adaption may reverse.

We advise all of you to read the report in full and make the necessary adjustments to your plans and pitches.

More Later

L Ravichandran


Great breakdown

Like
Reply

To view or add a comment, sign in

More articles by L Ravichandran

  • Two Articles on #IndustrialAI

    Two interesting articles on #IndustrialAI. The first one is from @Siemens on #IndustrialgradeAI and their Industrial…

  • AI Safety : Paper from Google

    I was pleased to see this research paper on AI safety solutions from Google Deepmind. The title of the paper is “ An…

    2 Comments
  • AI & Philosophy Blog 3 : Sciences moving in the opposite directions

    Preface: This blog is part of a series of blogs based on short essays compiled in the book “The Minds I” by Douglas R…

  • AI and Philosophy Blog2: Who are we?

    Preface: This blog is part of a series of blogs based on short essays compiled in the book “The Minds I” by Douglas R…

    1 Comment
  • AI and Philosophy Blog 1 : Are we one person or personality

    Preface: This blog is part of a series of blogs based on short essays compiled in the book “The Minds I” by Douglas R…

    2 Comments
  • A Bridge Too Far for "AI for Enterprise"

    Let us look at 3 emerging stories relating to enterprise adoption of AI. A fourth last story, we will keep for the end…

    1 Comment
  • Some thoughts on the latest GenAI news

    There is a lot of news lately on OpenAI losing large amounts of money in its operations and how opensource LLMs match…

    1 Comment
  • AI Agents and Prompt Engineering Thoughts

    AI Agents, prompt engineering and Systems Integration L Ravichandran AiThoughts.Org Many people thought that the…

    4 Comments
  • IT Outage Musings

    The lessons from the IT outage My favorite Gary Marcus raised a pertinent point: if a simple update of non-AI regular…

    1 Comment
  • AI Use Cases for Manufacturing Vertical

    AI Use Cases for Manufacturing Vertical Lots of news and talk about AI Usecases in Technology, retail consumer, and…

    3 Comments

Insights from the community

Others also viewed

Explore topics