Economic Benefits of Generative AI
We often look at the functional and operational benefits of AI, however, in most societies the economic benefits are the ones that create wealth, improve living standards beyond the technology, and whether it is right, judge the standard of success. In this article we shall look at the economic benefits of generative AI. I am not an economist so this will be very light touch.
The major source for this article comes from trade publications, Bing and Google searches, and software suppliers. Most of data and figures comes from McKinsey, (September, 2023) as I felt this could be a reliable and single source so as not to mix different measurement techniques.
Economic benefits of Generative AI
It is estimated that globally generative AI could add between $2.6 and $4.4 trillion annually economic value over the next decade. This is between 15% and 40% of the estimated $11 to $17.7 trillion that could be unlocked by all AI. The bulk of the impact would come from just five functions: customer operations, marketing/sales, software engineering, innovation, and R&D.
All industries are expected to have a positive impact from generative AI, but some specifics are:
Banking could see $200-340 billion in additional annual value
Retail/CPG $400-660 billion, and
Parma $60-110 billion.
The share of activities with automation potential has increased from 50% to up to 70% due to advances like natural language processing.
Economic boost from generative AI and other technologies, particularly from automation, could provide up to a 3% annual boost to productivity growth over the next 15 plus years. This relies on workers and businesses achieving the benefits and moving (staff and processes) to other productive activities.
Generative AI Enabling the Workforce
For customer operations, generative AI chatbots could be used to handle 50% more inquiries and to reduce response times. In marketing, it can generate creative personalised content at scale. For software engineering, tools like GitHub Copilot are enable faster coding. And in R&D, generative design can accelerate designing new products or drug molecules. Additionally, I think generative AI will be able to accelerate the creative processes for product design, and consumer testing.
Adoption scenarios suggest automation of half of current work activities could happen by 2060, a decade earlier than previous estimate of just 3 years ago. A difference is that unlike prior automation, generative AI is most likely to impact all wage groups.
Realising the Benefits
While promising, realising the benefits will take time. Business leaders need to address risks like biases, privacy issues, workforce impacts, and intellectual property infringement, while scaling in a controlled manner. Overall, if managed responsibly and carefully, generative AI could contribute significantly to economic growth and sustainability.
Some area a business can use generative AI to achieve economic benefits at an organisational level are:
Competitive Advantage: Organisations that adopt generative AI early can gain a competitive edge by delivering better customer experiences and more efficient services. Also increasing efficiencies, within workflow and data processing, cost savings by automating repetitive tasks,
Insights: AI can review large amounts of data to give insights into market trends, customer preferences, product data, and more, to inform better decision making.
New Business Models: AI can enable the development of new business models, such as subscription services or on-demand content creation.
Scalability: AI allows businesses to scale their operations more easily and cost-effectively.
Implementing AI into an Organisation
Implementing AI in a business can be a game-changer, but it requires careful planning. Here's a simple guide for businesses:
Preparation for AI Implementation
Start by educating the team about what AI is and how it can be used in your business/industry.
Define what you want to achieve with AI. This could include automating tasks, improving customer service, or enhancing decision-making, but it’s important to have goals.
Ensure you have high-quality data. As AI relies on data, clean, organized, and relevant data is a must!
Build a cross organisational team (or teams) with both business and technical expertise, including data scientists, machine learning engineers, and domain experts who understand your business.
Training will be critical to success. With generative AI, effort needs to go into business and people transformation, e.g. process redesign, change management, and upskilling.
What to Look For
Choose AI solutions that match your goals. This might be off-the-shelf software or customised solutions.
Make sure the AI system you choose can grow with your business.
Make sure the AI system can integrate with your existing tech stack.
Be aware of the ethical implications of AI. Ensure you follow ethical guidelines in your AI implementations. See below.
Dangers and Challenges
Mishandling customer data can lead to significant legal problems and damage your reputation. Ensure you have a robust data privacy policy and measures to enforce it.
AI models can inherit bias from training data. Regular auditing and retraining of models to mitigate bias should be a consideration.
In other posts we have discussed bias and data privacy but also think about intellectual property.
AI projects can be expensive. Monitor costs closely!
Short and Long Term Vision
Short-term projects can produce quick wins through automating repetitive tasks, however, these advantages may be temporary as competitors replicate them.
Ambitious long-term projects tailored to the business could provide deeper competitive advantages, but will likely require more exploration and perhaps uncertainty initially.
Gains from early short-term projects can help fund investment in the more complex longer-term initiatives.
Aim to balance short-term ROI from AI projects with longer-term, more ambitious projects that could provide competitive advantage.
Top Tips for Improving AI Success
Begin with a small, manageable project to learn from and build experience and confidence.
Encourage cross organisational teams to work on AI projects. The different viewpoints can help address potential blind-spots and dangers.
Test AI solutions in a controlled pilot environment before full deployment.
The AI world moves fast and you should provide support and encouragement for your team to stay updated with the latest advancements.
All employees will need some level of AI awareness. Invest in educating employees at every level on the AI basics. It is better to understand generative AI's capabilities and limitations rather than be afraid of it.
Implement strong testing practices for generative AI systems before deployment that can check for unfair biases, privacy issues, malicious uses, and more. These unwanted characteristics can be “trained out”.
Establish a feedback loop to improve AI systems based on real-world performance. This goes hand in hand with monitoring for bias etc.
Monitor the AI's performance and evaluate its impact on your business goals. It would be easy to “leave it to the AI” and this take things in the wrong direction.
Set clear guidelines and principles for using AI ethically and responsibly. Don’t be afraid to have conversations on bias, privacy, security, and transparency. And set out “red lines” that should not be crossed.
Give users control over how much they rely on AI tools vs human intelligence. Avoid over-automation.
Share best practices for AI across the organisation and look outside for industry practices.
Make AI ethics and fairness a leadership role. Senior executives should spearhead efforts to develop AI responsibly and serve as role models.
Implementing these practices can help maximise generative AI's benefits while minimising risks and negative impacts on people. The goal is to embed ethics into all stages of the AI lifecycle.
Related Articles
