Read on Twitter
Last week McKinsey released their generative AI report.
Here are the 10 key takeaways everyone must know:
Here are the 10 key takeaways everyone must know:
1/ Value potential by industry
Generative AI has the potential to generate $4.4 trillion in value across industries.
• $460 billion in high tech. Predominant factor: software engineering
• $390 billion in retail. Predominant factor: marketing and sales
Generative AI has the potential to generate $4.4 trillion in value across industries.
• $460 billion in high tech. Predominant factor: software engineering
• $390 billion in retail. Predominant factor: marketing and sales
2/ Value potential by function
4 business functions account for ~75% of the total annual value of AI:
• Customer operations
• Marketing and sales
• Software engineering
• Research and development
4 business functions account for ~75% of the total annual value of AI:
• Customer operations
• Marketing and sales
• Software engineering
• Research and development
3/ Key use cases
Banking→legacy code conversion
Retail→consumer research
Pharma→research and drug discovery
Example:
The University of Washington recently used machine learning for protein design.
It allowed them to tailor protein complexes to specific biological responses
Banking→legacy code conversion
Retail→consumer research
Pharma→research and drug discovery
Example:
The University of Washington recently used machine learning for protein design.
It allowed them to tailor protein complexes to specific biological responses
4/ Achieving human-level performance
AI is expected to match median human performance and reach top-quartile human performance earlier than expected.
McKinsey's estimate for AI's natural language understanding:
2017 estimate: 2027
2023 analysis: 2023
Timelines are shortening.
AI is expected to match median human performance and reach top-quartile human performance earlier than expected.
McKinsey's estimate for AI's natural language understanding:
2017 estimate: 2027
2023 analysis: 2023
Timelines are shortening.
5/ Automation is increasing
The total percentage of hours that could be automated by integrating technologies that exist today has increased from 50% to 60–70%.
The technical potential curve is steep because of the acceleration in generative AI’s natural-language capabilities.
The total percentage of hours that could be automated by integrating technologies that exist today has increased from 50% to 60–70%.
The technical potential curve is steep because of the acceleration in generative AI’s natural-language capabilities.
6/ Automation adoption has accelerated by a decade
McKinsey modelled adoption scenarios for the spent time on work activities reaching 50% automation:
2016 estimated midpoint was 2053
2023 estimated midpoint is 2045
This is an acceleration of almost a decade.
McKinsey modelled adoption scenarios for the spent time on work activities reaching 50% automation:
2016 estimated midpoint was 2053
2023 estimated midpoint is 2045
This is an acceleration of almost a decade.
7/ Generative AI is likely to have the biggest impact on knowledge work
Particularly activities involving decision-making and collaboration.
This previously had the lowest potential for automation.
Particularly activities involving decision-making and collaboration.
This previously had the lowest potential for automation.
8/ Automation has the greatest impact on higher-educated workers
Labour economists often note that the deployment of automation technologies has the most impact on low-skilled workers.
Generative AI has the opposite pattern—the greatest impact is with more-educated workers.
Labour economists often note that the deployment of automation technologies has the most impact on low-skilled workers.
Generative AI has the opposite pattern—the greatest impact is with more-educated workers.
9/ Displacing college degrees
High-wage knowledge work activities were previously considered immune from automation.
AI will challenge the attainment of a multiyear degree as an indicator of skill.
This could lead to a more skills-based approach to workforce development.
High-wage knowledge work activities were previously considered immune from automation.
AI will challenge the attainment of a multiyear degree as an indicator of skill.
This could lead to a more skills-based approach to workforce development.
10/ Propelling higher productivity growth
Global economic growth was slower from 2012 to 2022 than in the two preceding decades.
Generative AI helps accelerate productivity growth and compensates for declining employment growth by automating individual work activities.
Global economic growth was slower from 2012 to 2022 than in the two preceding decades.
Generative AI helps accelerate productivity growth and compensates for declining employment growth by automating individual work activities.
Don't just stay updated, stay ahead.
You can follow me @thealexbanks for more on AI.
If you liked this thread, you'll love the newsletter.
Subscribe here:
noise.beehiiv.com/subscribe
You can follow me @thealexbanks for more on AI.
If you liked this thread, you'll love the newsletter.
Subscribe here:
noise.beehiiv.com/subscribe
Help everyone learn and retweet this thread:
https://twitter.com/thealexbanks/status/1671854642484101121
Full report by McKinsey:
You can read more about AI protein design here:
https://t.co/8tT9Qjg42bmckinsey.com/capabilities/m…
You can read more about AI protein design here:
https://t.co/8tT9Qjg42bmckinsey.com/capabilities/m…
https://twitter.com/thealexbanks/status/1652285361656324097
• • •
Missing some Tweet in this thread? You can try to
force a refresh
