Machine learning statistics

Machine learning statistics

August 9, 2022

Tatyana Korobeyko

Data Strategist

Machine learning (ML) is a subset of artificial intelligence (AI), which is defined as an ability of a machine to use historical data and algorithms to imitate how humans learn, gradually increasing its accuracy. Relying on machine learning consulting services, companies across every industry deploy ML-based solutions to improve productivity, decision-making, product and service innovation, customer journey, and more.   

These days, it’s not an exaggeration to say that each of us encounters machine learning multiple times each day – mobile banking, top news pop-ups, recommended content in social networks, Uber’s commute estimations, chatbots – the list goes on. And even though saying that machine learning entirely changes the way we live may sound like a cliché, we’re going to say and prove that once again with the most impressive and recent statistics. Here, we’ve compiled a list of stats and facts about ML and AI market share, use cases, adoption across industries and business functions, ongoing investments, talents, and more.

Machine learning market and adoption rate statistics

The global machine learning market is steadily growing: in 2021, it was valued at $15.44 billion, and owing to the increasing adoption of technological advancements, it is expected to grow from $21.17 billion in 2022 to $209.91 billion by 2029, at a CAGR of 38.8%. (Fortune Business Insights).

  • $432.8 billion is the forecasted worldwide revenues for the artificial intelligence market, including software, hardware, and services in 2022. (IDC)
  • In 2023, the AI market is expected to reach the $500 billion mark, and in 2030 - $1,597.1 billion (with a registered CAGR of 38.1% from 2022 to 2030). (IDC, Precedence research)
  • $24.9 billion is the projected worth of the AI market in North America in 2022. (Statista)
  • 35% of companies report using AI in their business, and an additional 42% of respondents say they are exploring AI. (IBM)
AI adoption rates around the globe
  • Larger companies are twice as likely to have actively deployed AI, while smaller companies are more likely to be exploring or not pursuing AI and all. (IBM)
  • 46% of organizations are planning to implement AI in the next three years. (Deloitte)
Emerging intelligent automation technologies
  • By 2025, nearly 100% of enterprises will be implementing some form of AI. (Forrester)
  • The three top drivers of AI adoption are:
    • The technology’s increasing accessibility
    • The need to reduce costs and automate key processes
    • The increasing implementation of AI into standard off-the-shelf business applications (IBM)
  • 1 in 4 companies is adopting AI because of labor or skills shortages. (IBM)
ML and AI adoption drivers

Looking for a trustworthy machine learning consultancy?

Contact us

Machine learning use cases

Although today's use cases for machine learning are becoming more varied, customer-centric applications remain the most common. According to Statista, 57% of respondents state customer experience represents the top ML and AI use cases.

ML and AI use cases for companies worldwide in 2021

Machine learning for marketing and sales

  • 49% of organizations using ML and AI in their marketing and sales processes, whether currently in production or in pilots, apply it to identify sales prospects, and 48% - to gain insight into their prospects and customers. (Harvard Business Review)
  • 67% of respondents to a recent survey from Harvard Business Review Analytic Services agree that ML and AI in marketing and sales will be critical to their company’s ability to compete in the future. (Harvard Business Review)
  • 31% of respondents who are using ML and AI in sales and marketing say they have increased their revenue and market share. (Harvard Business Review)
  • 66% of marketing leaders state their teams can focus more on strategic marketing activities due to automation and machine learning. (Think with Google)
  • 78% of marketers feel enthusiastic about deploying automation and machine learning as it gives them more time to focus on their priorities. (Think with Google)
  • More than 50% of leading performance agencies shifted more than 30% of their time to strategic activities thanks to machine learning. (Think with Google)
  • 7% of respondents see a fear of job loss among sales and marketing teams from the use of AI and automation. (Harvard Business Review)

Machine learning for healthcare

  • The three areas with the biggest AI potential in healthcare are ML-based medical diagnosis, early identification of potential pandemics and tracking incidence of the disease, and imaging diagnostics (radiology, pathology). (PwC)
  • The global AI in the healthcare market was valued at $11.06 billion in 2021 and is expected to reach $187.95 billion by 2030. (Precedence Research)
  • The clinical trials segment dominated the global AI in the healthcare market and brought over 24.2% of revenue in 2021. (Precedence Research)
  • North America led the AI in the healthcare market in 2021 and is forecasted to continue doing so up to 2030. (Precedence Research)
  • 70% of drug discovery costs can be cut with the appliance of AI and ML.  (Insider Intelligence)
  • Up to 95% of accuracy in predicting Covid-19-related physiological deterioration and death up to 20 days in advance can be achieved with an ML-based solution.  (Nature)

Machine learning for banking

  • According to McKinsey, AI and its related technologies will have a seismic impact on all aspects of the insurance industry, from distribution to underwriting and pricing to claims. The biggest ML and AI potential in the sector will be observed in the areas such as personalized financial planning, fraud detection and anti-money laundering, and process automation. (PwC)
  • The global AI in banking market size was valued at $3.88 billion in 2020 and is projected to reach $64.03 billion by 2030, growing at a CAGR of 32.6% from 2021 to 2030. (Business Wire)
  • 80% of banks are highly aware of the potential benefits presented by AI and machine learning and 75% of respondents at banks with over $100 billion in assets are currently implementing AI strategies.  (Business Insider)
  • 60 % of financial-services sector respondents state they have already embedded at least one AI capability. (McKinsey)
  • Automating middle-office tasks with ML and AI can save North American banks $70 billion by 2025. (Insider Intelligence)
  • $447 billion is the estimated aggregate potential cost savings for banks from ML and AI applications by 2023, with the front and middle office accounting for $416 billion of that total. (Insider Intelligence)
  • AI platform revenues within insurance are forecasted to grow by 23% to $3.4 billion between 2019 and 2024. (GlobalData)
  • 56% of insurance executives believe ML and AI will significantly improve operational efficiency over the next three years (2020-2023). (GlobalData)
  • One of the biggest roadblocks keeping banks from executing their ML and AI strategies is the “black box” dilemma. (Deloitte)
  • 76 % of respondents of Statista’s report consider applying AI and ML technology in stock market workflows. (Statista)

Machine learning for manufacturing

  • The three areas with the biggest ML and AI potential in manufacturing are auto-correction of manufacturing processes, supply chain and production optimization, and on-demand production. (PwC)
  • The US AI in manufacturing market size was estimated at USD 543.42 million in 2021, expected to reach USD 788.82 million in 2022, and is projected to grow at a CAGR of 45.91% to reach USD 5,245.50 million by 2027. (ReportLinker)
  • 93% of companies believe AI will be the key growth and innovation driver, the Deloitte’s survey on AI adoption in manufacturing states. (Deloitte)
  • 83% of companies consider that ML and AI will have a tangible effect on manufacturing in two to five years. (Deloitte)
  • Only 9% of manufacturing companies state ML and AI projects meet their expectations in terms of achieved benefits, budget and time invested, etc. (Deloitte)

Machine learning for retail

  • According to the PwC report, the three areas with the biggest ML and AI potential in retail are personalized design and production, anticipating customer demand (e.g., using predictive analytics tools to foresee customers’ orders in advance), and inventory and delivery management. (PwC)
  • The global AI in the retail market is forecasted to grow from $4.84 billion in 2021 to $31.18 billion at a CAGR of 30.5% in 2028. (Fortune Business Insights)
  • Retail will remain the largest AI spending industry in the US in 2021-2025. (IDC)

Machine learning benefits

Being an extension, not a replacement for human capabilities, machine learning enables companies to automate complex processes, improve the quality, effectiveness and creativity of employee decisions with rich analytics and pattern prediction capabilities, and uncover gaps and opportunities in the market to introduce new products and services, hyper personalize customer experience, and much more. (Accenture)

As ML and AI initiatives are becoming more widespread, companies are getting more value out of their investments, as you can see from the following numbers and facts: 

  • 30% of global IT professionals claim their employees are already saving time with new AI and automation software and tools. (IBM)
  • 31% is an average cost reduction expected by organizations adopting intelligent automation in the next three years. (Deloitte)
Intelligent automation benefits
  • By 2024, AI-powered enterprises will respond 50% faster to customers, competitors, regulators, and partners than their peers. (Oracle)
  • According to the study requested by the Committee on Industry, Research and Energy (ITRE), AI and machine learning will contribute to the labor productivity increase up to 37% by 2025. 
  • 42% of companies stated that the profitability of their ML and AI initiatives exceeded their expectations, while only 1% said it didn’t meet expectations. (Accenture)
  • 92.1% of companies state they are achieving returns on their data and AI investments. (NewVantage Partners)
  • 27% of the respondents of the latest McKinsey Global Survey on AI report at least 5% of earnings before interest and taxes attributable to machine learning-based AI.
  • PwC research shows that the global GDP could be up to 14% higher in 2030 (up to $15.7 trillion1 to the global economy) as a result of the ML and AI accelerated development and take-up. (PwC)
  • According to PwC research, 45% of total economic gains by 2030 will be the result of AI-driven product enhancement, stimulating consumer demand.
Key drivers of the AI economic impact
  • All regions of the global economy will benefit from ML and AI, with China and North America seeing the biggest economic gains with AI enhancing GDP by 26.1% and 14.5% accordingly. (PwC)
Estimated economic gains by region by 2030

Investments in ML

With such inspiring benefits, no wonder enterprises are increasing their investments in ML and AI initiatives. For example, solely in the United States  the spending on artificial intelligence will grow up to $120 billion by 2025, representing a compound annual growth rate (CAGR) of 26.0% in 2021-2025. (IDC)

  • In 2021, 59% of respondents stated that accelerating investments in AI and machine learning is a part of a strategy for becoming future-proof to changing customer demand and operational challenges. (Statista)
  • 91% of top businesses report having an ongoing investment in AI and ML and 91.7% said they are increasing their investments. (NewVantage Partners)
  • 96% of survey respondents plan to use AI simulations, such as digital twins, in 2022.  (PwC)
Plans to use AI simulations
  • The US industries that will see the fastest growth in AI spending are Professional Services, Media, and Securities and Investment Services, with CAGRs greater than 30%. (IDC)
  • Retail and banking will represent nearly 28% of all AI spending in the United States in 2025. (IDC)

Machine learning skills demand and employment statistics

While AI and ML are becoming mainstream, the advances in AI and ML are being slowed by the shortage of employees with required skills. According to Statista, 82% of organizations need machine learning skills and only 12% of enterprises state the supply of ML skills is at an adequate level.

Demand and supply for AI-related skills in enterprises worldwide

This shortage has the potential to hold back digital innovation and economic growth. According to the IBM Global AI Adoption Index in 2022 research, 34% of organizations consider insufficient AI skills, expertise or knowledge as the top reason blocking successful AI adoption. (IBM)

Convinced that with an effective AI team in place they would leverage AI to its full potential, companies are now allocating their budgets to:

  • Retraining and upskilling existing employees - 68% of companies
  • Identifying and recruiting skilled talent from other companies and organizations - 58% of respondents
  • Recruiting from universities - 49% of companies (SnapLogic)

Struggling to find specialists with adequate AI and ML knowledge, organizations put focus on such hard skills as:

  • Coding, programming, and software development – 35% of companies
  • An understanding of governance, security and ethics – 34% of companies
  • Data visualization and analytics – 33% of companies
  • Advanced degree in a field closely related to AI/ML – 27% of companies (SnapLogic)
Skills and attributes organizations are looking for in their AI teams

As for soft skills needed for these tech roles, 37% of the respondents of the IBM’s survey Addressing the AI Skills Gap in Europe consider problem-solving to be the most critical soft skill and 23% of tech recruiters struggle with finding applicants with this aptitude. (IBM)

Other facts and statistics:

  • Data scientist is one of the in-demand jobs employers find hard to fill. (Indeed)
  • The average salary of an ML engineer varies from $112,342 to $145,688, with the final number impacted by experience, industry, and geographic location. (Coursera)
  • $124,052 is the estimated total pay per year for a machine learning engineer in the US, with an average salary of $100,321 per year.  (Glassdoor)
  • The employment of machine learning engineers is projected to grow by 22% between 2020 and 2030. (US Bureau of Labor Statistics)
  • 40% of tech job seekers and employees consider software engineering and knowledge of programming languages to be the critical technical capabilities for the AI/tech workforce. (IBM)
  • 73% of the information technology decision-makers (ITDMs) in the UK and 41% of ITDMs in the US lack the right in-house AI talent to execute their AI and ML strategy. In both the US and UK, manufacturing and IT are challenged the most by this in-house talent shortage. (SnapLogic)
  • 97 million new jobs across 26 countries will be created with the appliance of AI by 2025. (World Economic Forum)

Power your digital transformation with machine learning

Machine learning consulting services

Machine learning consulting services

Itransition’s ML team builds and implements custom ML-powered solutions as well as helps you optimize the existing ML workloads and scale ML across the enterprise.

Machine learning challenges

Based on the findings of IDC research Thrive in the Digital Era with AI Lifecycle Synergies, 50% of the AI initiatives fail and only one-tenth of AI and ML PoCs reach production deployments. IDC quotes the lack of data science skilled personnel, the cost of AI solutions, data management issues (data quality, quantity, and access), and issues with the ML algorithm explainability and selection as the main reasons for AI initiatives’ failure. (IDC)

According to some sources (Statista, IBM), the inability to scale AI and ML projects up tops the challenges that are hindering AI from reaching its potential.

Machine learning challenges
Top 5 AI challenges

Even though 85% of IT professionals agree that consumers are more likely to choose a company that’s transparent about how its AI models are built, managed and used, the IBM Global AI Adoption Index 2022 research reveals that a majority of organizations haven’t taken key steps to ensure their AI is trustworthy and responsible:

AI adopters' struggles

Other facts and statistics:

  • 94% of AI systems used to scan for the signs of breast cancer were less accurate than the analysis of a single radiologist. (British Medical Journal)
  • 56% of respondents experienced issues with security and auditability requirements when deploying ML and AI tools in 2021. (Statista)
  • Some health-tech firms providing AI healthcare solutions don’t keep patient data 100% confident. (Bloomberg)
  • Even though in some cases ML models seem effective for event-level prediction of crimes (90% of accuracy was achieved across 8 US cities), some studies and research prove the opposite. For example, only 19% of the computer-generated matches turned out to be correct when Metropolitan Police officers used a face-recognition system to look for suspects on the streets. (Bloomberg)

A closing word

As seen from the statistics above, each company regardless of the industry has an endless number of ML adoption scenarios and high chances of success in case of following through on their initiative. If you want to advance your ML usage and achieve tangible economic gains, you need to take a holistic approach to AI and ML adoption. Rather than focusing on implementing scattered ML-powered solutions to address specific business needs, think of ML as an enabler of business transformation, enhanced decision-making, and modernized systems.

Machine learning in ecommerce: 9 use cases reshaping online shopping