The Ultimate List of Machine Learning Statistics for 2023

Updated Jan 10, 2023.

Machine Learning (ML) has grown in popularity thanks to the innovation of exciting products by the industrial and academic sectors. As a result, ML has become a hot topic among various stakeholders.

Applications like artificial intelligence software and business intelligence software are great examples of tools that use machine learning algorithms to understand customers and increase conversion rates.

If you're wondering how you can take advantage of the technology, sit tight because we'll cover the latest machine learning statistics that will help you better understand the topic and maybe even profit from it.

Key Machine Learning Statistics

  • The most profitable area of machine learning is in marketing and sales. (McKinsey)
  • 38% of jobs in the USA can be automated by 2030. (PwC)
  • Cost reduction is the number one reason why companies use machine learning. (Algorithmia)
  • You can increase customer satisfaction by around 10% with the help of machine learning. (Forbes)
  • Nissan used a machine learning model to increase their conversion rates by 67%. (Think With Google)

Machine Learning Adoption Statistics

1. Experts forecast the global machine learning adoption rate to be around 42% CAGR between 2018 and 2024.


Small to large businesses alike are adopting machine learning on a massive scale, as 65% believe the technology will help them analyze data and make better decisions.

The latest ML stats show that Europe and North America lead the ML adoption race at 44.87% and 44.05%. Asia, Africa, Oceania, and South America make a combined 11.08% market share.

With so many positive reviews about the thousands of artificial intelligence tools flooding the market, it's safe to say if your company is not leveraging data analytics and insights from ML, you'll have a tough time competing against those who do.

Global Mashine Learning Market share

2. 91.5% of leading businesses have ongoing investments in ML and AI.


ML is the most critical aspect that allows different artificial intelligence applications and technologies to function or solve problems.

These include speech and pattern recognition, standard deviation observations, regression analysis, robotics, and predicting stock market highs and lows, to name a few.

The most common challenges businesses face when adopting AI are scaling up (43%), versioning machine learning models (41%), and getting senior buy-in.

3. Advancements in machine learning and artificial intelligence may increase the global GDP by 14% by 2030.

(WSJ, McKinsey)

The total funding for ML in the first quarter of 2022 was $29 billion throughout the world.

However, descriptive statistics reveal that budgets for machine learning projects will increase by 25%, with information technology, manufacturing, and banking industries experiencing the highest growth.

In a study conducted by McKinsey, 50% of respondents said they adopted AI and ML into at least one business function.

Does your current employer incorporate machine learning methods into their business

4. 44% of respondents said they deployed ML in pockets.


10% of the respondents said they are still experimenting and investing in people and infrastructure. In the same study, at least 20% of C-level executives from 14 industries reported machine learning as the most vital part of their business.

As more and more organizations become aware of the benefits of machine learning models, we’ll start to see these C-Level executives using the technology for reporting performance data analytics.

Average number of AI or ML Projects Deployed

5. Using machine learning rarely decreases business expenses, but it surely does increase your revenue, as reported by 80% of people in a survey by McKinsey.


One might expect fundamental cost decreases when leveraging machine learning techniques and statistical models, but instead, revenue is the one that increases. While this isn’t a bad thing, it might surprise some who thought the two factors had to happen simultaneously.

6. 25% of top IT leaders believe machine learning programs will help curb security risks in their respective companies.


Security is a grave concern for most businesses. Even though machines get smarter by the day, hackers often always find new ways to outmaneuver them. However, machine learning methods seem to help with this issue.

Biggest reasons for machine learning technology adoption in organizations worldwide

7. 33% of top IT leaders adopt machine learning mainly to improve their business analytics processes.


ML is essential for business analytics because businesses often generate terabytes of real-world data points per second.

That is why applications that leverage machine and statistical learning algorithms are flooding the market today, as they can help business owners better understand big data.

Successful companies applying data extensively are 23x more likely to acquire new customers than laggards (source: big data statistics list).

Machine Learning Use Cases

8. Cost reduction is the number one reason why companies use machine learning.


A recent study shows that businesses with over 1,000 employees use ML for internal applications, such as automating internal processes or reducing operating costs.

Smaller businesses use statistics and machine learning to generate customer insights and enhance customer experience.

Overall, the study concludes that the bigger a company becomes, the more they optimize sales and prioritize cost-saving measures. They start to shift their focus from customer service to improving product lines.

The Leading Machine Learning Use Cases

9. According to C-level Executives and Data Scientists, the top use of ML is Risk Management (82%).

(Refinitiv, MemSQL, Statista)

The second most popular use case of machine and statistical learning is Performance Analysis and Reporting (74%), followed by:

  • Trading Investment Idea Generation (63%),
  • Automation (61%).
  • Business Analytics (33%)
  • Security (25%)
  • Sales & Marketing (16%)
  • Customer Service (10%)
  • Other (16%)

10. 14% of the organizations surveyed took 0 to 7 days to deploy a single machine learning model.


  • 28% took 8 to 30 days.
  • 22% took 1 to 3 months.
  • 13% took 3 months to 1 year.
  • 5% took more than a year.
  • 18% are not sure or don’t know about the implementation time of a machine learning model.

11. 51% of respondents from a survey by O’Reilly say they use internal data analysts and machine learning engineers to build their ML models.

(Algorithmia, O’Reilly)

On the other hand, AutoML services from cloud providers are low – a split that is even clearer among advanced ml users. 12% of those who are part of organizations but with less background knowledge and experience with the technology rely on external consultants for developing models.

However, as more and more businesses see the importance of model-based performance analysis and data collection, that number will increase considerably.

Who builds the ML models

12. An ML algorithm can detect if a child has epilepsy with 73% accuracy.


This rate is almost similar to that of adults, but it's more accurate in children as their brains are still developing. Most doctors see this as a huge breakthrough and hope it’s incorporated into clinical use sooner than later.

13. A machine learning algorithm can detect lung cancer with 94% accuracy.

(Health IT Analytics)

In the USA, lung cancer is the most common cause of cancer death. Even though early detection reduces the mortality rate by 45%, the equipment-related challenges, lengthy procedures, and high costs make early diagnosis difficult.

Because of that, data scientists are starting to adopt deep learning systems as a better solution. The model reduces false negatives and positives by 5% and 11%, which is better than six human radiologists.

14. A deep learning model can predict under-performing businesses with an accuracy of 62%.


Last year, financial experts and Microsoft engineers gathered empirical data to try and automate the preliminary review process of investment documents.

They then developed a deep learning model that uses a one-dimensional neural network based on text from public financial statements of any company.

15. A machine learning algorithm can read lips with 95% accuracy.


That is one of the biggest wins of machines over humans as the algorithm performed better than experienced human lip readers, who only averaged a 52% accuracy.

The algorithm operates in the recurrent neural network (RNN), which interconnects with the artificial neural networks that can process different sequences of inputs.

These networks produce accurate predictions and are highly effective in various applications, from speech recognition to handwriting and texts.

16. Machine learning facts show that human editors can only tell the difference between AI and human authors 50% of the time.

(The Conversation)

GPT-3, the latest and most comprehensive text modeling system, has over 175 billion parameters. It recently published a piece in the Guardian newspaper trying to convince us that it’s not looking to dominate the world. Whether that’s true or not is up to you to decide.

17. After Nissan started using Google’s automated bidding products, its conversion rates increased by 67%.

(Think with Google)

ML facts reveal that Nissan optimized their ads in real-time, which gave them control of where to display the ads and constantly calibrate customer targeting.

On top of their increased conversion rates, the cost per qualified visit decreased by 33%. Most top company leaders plan to use ML marketing solutions to develop models, improve their profits, data mining, and optimize campaigns.

18. The accuracy of predicting a patient's death by Google AI is 95%.

(Bloomberg, Google AI Blog)

Google’s deep learning machine learning program is also 89% accurate in detecting breast cancer.

That makes it 15% more effective if you compare it to pathologists, who are accurate only 74% of the time. The accuracy of predicting death for a COVID-19 patient is 92%.

When it comes to translation accuracy, Google Translate went from 55% to 85% after they started using a machine and statistical learning algorithm.

19. Amazon uses a machine-learning algorithm to automate picking items and packing them in logistic regression.


With the help of regression trees and Kiva’s capabilities, the average ‘click to ship’ time went down by 225%. That is around 60-75 minutes to 15 minutes.

20. Netflix saved over $1 billion because of its machine learning system.


A stat from Bloomberg shows that 80% of the movies people watched on Netflix were recommended to them by their reinforcement learning algorithm. Because of this, Netflix has planned to put aside a whopping $19 billion for programming.

Such a huge investment to help serve more personalized recommendations to users comes after the rise of competitors like Hulu, HBO Max, Amazon Prime, Peacock, and Disney+.

The same algorithmic capability you see on Netflix is the same one on popular platforms like TikTok, YouTube, and Instagram. Now that ML use is the standard of how businesses operate, it may soon be as accessible as our smartphones.

Netflix - System overview Recommendation involving multi-layered machine learning

21. Facebook has an AI-powered face recognition model with 97% accuracy.

(Facebook Research, BBC)

In 2015, a research group turned Facebook engineers developed a neural network system called DeepFace. Although industry experts praised the system, it didn’t sit well with many as it added fuel to the fire on the rising privacy concerns. After launching the system, Facebook's CEO faced a class-action lawsuit.

22. Tesla recorded over 4 billion autonomous miles as of Jan 2021.


Machine learning is responsible for the rise of applications found in self-driving cars. The technology enables artificial intelligence to train the computer in cars to learn without human intervention. As we speak, developers continue to innovate ways machines learn, like through:

  • Statistical learning
  • Reinforcement learning
  • Supervised learning
  • Unsupervised learning
  • And various other machine learning tasks
Estimated Tesla Autopilot Miles

Machine Learning Market Statistics

23. The ML market is forecasted to grow from $1 billion in 2016 to $9 billion in 2022, at a CAGR of 44%.


In 2021, the global machine learning market had a value of $8 billion, and experts say it will reach around $75.54 billion by 2023 and $117 billion by 2027. As the market grows, new computer science statistics show that its most prominent segment (the deep learning software category) will reach a market cap of over $1 billion by the year 2025.

Additionally, AI-powered assistants and hardware will also experience unprecedented growth – hitting a valuation of $87.68 billion, at a CAGR of 39% from 2022 to 2026. This growth can deliver $13 trillion to global economic activity by 2030.

24. The US deep learning and machine learning market will be $80 billion by 2025.


Even though this number is already huge, it will increase as businesses leverage ML algorithms.

According to descriptive statistics from McKinsey, deep learning techniques such as convolutional neural networks, recurrent neural networks, and feed-forward neural networks are responsible for 40% of the annual value of all analytical methodologies.

25. COVID-19 is to blame for the production decrease of 12% in the ML chip-making business.

(Market Data Forecast)

The pandemic has also decreased the overall sales of chips by 13%. Surprisingly, many thought the drop would be significantly higher than this.

Careers in Machine Learning

26. There are less than 10,000 individuals who have the necessary skill to tackle serious AI problems.

(Accenture, O’Reilly)

According to O'Reilly, you can find jobs dedicated to machine learning in firms with a lot of expertise in ML, such as deep learning engineers (20%), machine learning engineers (39%), and data scientists (81%).

27. On LinkedIn, data scientists rose by over 650% between 2012 and 2021.

(Towards Data Science)

On, the most in-demand talents are natural language processing (NLP), deep learning, and machine learning.

28. Only 4.5% of data researchers or data scientists in the USA work exclusively as ML engineers.


18% of companies in the USA had 11 or more data scientists on their payroll in 2018. However, the number grew by 39% to 15 or more in 2021.

That shows that businesses are increasing their hiring efforts to expand their data science teams. According to Kaggle, the median annual income of a full-time data scientist in the USA was $120,000 in 2021.

Machine Learning in Marketing

29. 16% of respondents from a Refinitiv survey said they believe ML would improve their sales and marketing campaigns.


Several businesses are starting to use machine learning algorithms for targeted marketing, proving to be much more efficient than the old-school advertising tactics.

30. 61% of marketers claim that AI is the most vital part of their data strategy.

(Inside Big Data).

As customers' expectations have changed, you'll need to communicate personally to get their attention to effectively market your service or product.

That is why 56.5% of marketers use artificial intelligence and machine learning to personalize their content, run intelligent campaigns, and create a great user experience for customers.

31. 87% of AI adopters say they use or consider using it for sales forecasting or improving their email marketing campaigns.

(Venture Harbour)

As we all know that marketing is a resource- and labor-intensive process, the use of ML for sales forecasting and email marketing by marketers should not come as a surprise.

ML can help speed up customer churn prediction, improve lead scoring accuracy, and create dynamic pricing models, among many other things. Because of using ML automation and technologies for marketing purposes, marketers are two times more likely to report a positive ROI than those without the technology.

32. Marketing and sales are the most profitable departments to incorporate machine learning systems.


ML allows you to personalize algorithms to generate product recommendations tailored to the site visitor. You derive these algorithms from various data sets on consumer behavior that help increase customer retention and sales conversion.

Average Revenue Increase from AI Adoption

Machine Learning in Business

33. Businesses leveraging machine learning algorithms can increase business productivity by 54%.


In simpler terms, ML is helping employees to make better use of their time, leading to increased revenue generation and profit.

34. C-level executives are usually responsible for overseeing 75% of AI projects in their respective firms.


Some years ago, executives knew nothing about machine learning. Not even how ML algorithms impacted their firms.

Fast-forward to the present time – the situation has flipped drastically due to the technology hype. Top executives are now leading the firm's AI and ML developments. They use AI to cut out redundant tasks like timesheets (78%), scheduling (79%), and paperwork (82%).

A report by McKinsey forecasts that investment in artificial intelligence and ML will increase by over 300% in the coming years. So if you're looking to become an executive anywhere, make sure you equip yourself with knowledge on these topics.

35. Customer service is synonymous with artificial intelligence. That's why over 80% of businesses will use it in this department.


With clients wanting more value from interactions with businesses, you need to adjust to keep up with the current state of the market – and what better way than with the help of ML algorithms.

AI helps automate parts of the customer service process, which causes the customer experience to be more worthwhile. Recent data science stats from B2C shows that 45% of customers prefer communicating with chatbots if they have any customer service inquiries.

What's more astonishing is that 62% of the respondents from the same study were willing to submit personal data to the AI bot to get a better user and business experience.

36. 15% of all manufacturing businesses say they are willing to use AI in widespread production.


Only a few manufacturing companies worldwide use AI to produce products. Even though the number may seem low, machine learning is a relatively new development – the number is still more than the expected value.

37. There were over 4400 funding rounds for machine learning businesses in 2021, raising $73.7B from all the rounds.

(Crunchbase, Forbes)

Because of this, startups are looking for new, skilled talent. We see over 44,000 jobs on LinkedIn in the US and 98,000 jobs worldwide requiring the applicant to have a skill in machine learning. So if you're serious about upgrading your skills, ML should become a priority on your list.

Machine learning tops AI funding worldwide

38. 75% of businesses that use AI and machine learning increased customer satisfaction by around 10%.


Inferential statistics estimate that 32% of companies are starting to invest more in customer service to improve customer satisfaction. Data mining and automation allow marketers to perform tasks more efficiently, which helps them satisfy and retain customers longer.

Machine Learning Methods for Sales Teams

39. Businesses that use AI for sales can increase their leads and appointments by over 50%.


They also reduced call time by between 60 and 70% and had cost reductions of between 40 and 60%.

Recent computer science statistics show us that there's been an increase in the number of sales teams taking to ML to assist them in moving customers faster through the sales funnel.

Slowly but surely, it’s becoming one of the most crucial sales enablement solutions for higher conversions. With the help of AI, 34% of buyers are willing to spend more, while 49% shop more frequently.

Machine Learning in Voice Assistants

40. By 2026, the value of the global natural language processing market will be around $42.04 billion.

(Mordor Intelligence)

NLP uses deep and statistical learning. Two closely related fields to ML because you use training data to build the models.

This technology is behind voice assistants like Google Assistant, Echo, and Siri. Because of the mobile technology boom, the popularity of voice assists will only rise, as we see from the above ML statistic.

41. 50% of people around the world use voice assistants.


That's around 3.25 billion people using voice assistants and voice-activated searches worldwide.

Forecasts also reveal that by 2023, close to 8 billion people will be using voice assistants.

As they become widely known and incorporated within households, businesses that bet on AI will start reaping the rewards of their investments.

As people want to increase their productivity, getting tasks completed through voice commands is a step in the right direction for most.

42. During the COVID-19 pandemic voice, AI usage increased by 7%.


The global COVID 19 pandemic seems to have positively impacted the usage of voice assistants, as the number of times people used it per day rose to 25% between March and April 2020.

Between December 2019 and January, the daily usage stood at only 20%. This data set is crucial for businesses on the fence about incorporating ML in their voice assistants or taking their voice-assisted capabilities to a higher level.

43. In 2022, using voice assistance multiple times a day went up by 5% in the past six months.


People never used to use voice commands. According to a survey by Voicebot, respondents said they now use it multiple times a day.

Because of this, businesses should add voice assistants to their business plans to help customers interact with their products and services in a more user-friendly way.

Businesses should ideally do this with the assistance of machine learning, which adapts to the needs and behavior of customers over time to provide them with personalized service.

44. 80.5% of consumers under 30 use a voice assistant on their smartphones versus 60.5% of those over 30.


If we dig a little deeper into the machine learning statistic from Voicebot, we see that 74.7% of consumers between the age of 30 and 44 use voice assistants on smartphones, compared to 68.8% of consumers between the ages of 45 to 60.

Voice assistants are gaining wider adoption

General Machine Learning Statistics

45. 59% of professionals say Tensorflow is their favorite ML platform.

(Statista, Forbes)

Tensorflow is an open-source deep and supervised learning framework developed and released for commercial use in 2015 by Google Brain. The service has grown because of its adaptability and flexible architecture. LinkedIn's Newsle (88.86%) has the largest ML software market share, followed by Tensorflow (3.38%).

Market Share of the Leading Machine Learning Software

46. 38% of jobs in the US can be automated by 2030.


The US leads other top markets like Germany (35%), the UK (30%), and Japan (21%). Sectors of the economy that are at the highest risk of automation are storage and transportation (56%), manufacturing (46%), and retail and wholesale (44%). Sectors at low risk are social work and healthcare (17%).

Share of tasks performed by humans vs machines, 2022 and 2025 (expected)

47. Employment in the research and computer science space is forecasted to grow by 15% from 2019 to 2029.


This rate is higher than the average for other occupations (4%). With the machine learning market growing exponentially, occupations relevant to it will be in abundance and likely be very lucrative.

48. Of the newly filed patent applications in the healthcare industry, 40% have an ML or AI aspect.

(Kluwer Patent Blog)

Computer science stats show that the European Patent Office has started to receive patent applications for ML or AI applications frequently. In response, they‘ve now changed their guidelines to allow applicants to use these programs.

EPO now treats machine learning and AI as mathematical methods, which are not patentable because they consider them non-inventions. But if a mathematical methodology controls a technical process or system, it becomes patentable as it has gained a technical character.

Machine Learning FAQs

What’s the difference between machine learning and AI?

Although used interchangeably, machine learning is closely related to AI. Artificial intelligence is a much broader topic that encompasses machines carrying out tasks in an ‘intelligent way' as they are programmed to mimic humans.

Will machine learning replace jobs?

Yes, to a certain extent. ML and other subsets of AI help eliminate redundant jobs like proofreading, bookkeeping, and toll collecting. Even though that may be the case, the AI revolution will create more jobs, and humans with the required skill will be able to coexist with machines.

When does machine learning fail?

Machine learning fails when training data is of poor quality. It deducts nuances and patterns from data input to gain insight and enhance performance. If data is not enough, the algorithms' predictions will be lackluster. 

Where is machine learning used?

You can use ML in different industries to address various challenges. Good examples are fraud management companies, stock markets, and elderly care services.

Grow Your Business With Machine Learning

Having read the above machine learning statistics, you'll better understand what machine learning is and how it can impact your life and business.

In a nutshell, ML’s main aim is to empower people using it to make calculated predictions based on input data.

And to do that effectively make use of data in your business, you’ll need the best business intelligence tools:

  • Zoho Analytics allows you to upload and connect to various data sources in different formats. This tool has a drag-and-drop interface, which you can use to combine reports and make interactive graphs and charts.
  • Big Eval is a bit more user-friendly than Zoho Analytics, plus it allows you to perform in-depth quality checks by testing algorithms, case execution, and case organization. This tool has a free demo where you'll be walked through all the features and have your questions answered by an expert.
  • Yellowfin has a free 30-day trial plan and a mobile app – unlike the other two. It can help you automatically analyze data and share information to workflows for reuse in storyboards, stories, and dashboards.

Businesses and individuals in various disciplines worldwide are starting to embrace machine learning. Don’t be left behind. Get started today!


  1. McKinsey
  2. Market Research Future
  3. Statistics and Machine Learning Toolbox
  4. Statista
  5. Forbes
  6. Emerj
  7. Algorithmia
  8. Think with Google
  9. Forbes
  10. Statista
  11. Facebook Research
  12. Built In
  13. Statista
  14. BBC
  15. Forbes
  16. Refinitiv
  17. B2C
  18. McKinsey
  19. Review42
  20. AUM
  21. Fortune
  22. WSJ
  23. Bloomberg
  24. SAS
  25. Oberlo
  26. Salesforce
  27. Business Insider
  28. Accenture
  29. Venture Harbour
  30. Businesswire
  31. Fortune Business Insights
  33. Harvard Business Review
  34. Statistics and Machine Learning Project
  35. MIT
  36. PWC
  37. Health Analytics
  38. Inside Big Data
  39. VB

Was This Article Helpful?

Rated 5 out of 5
5 out of 5 stars (based on 1 review)
Very good0%

Martin Luenendonk

Editor at FounderJar

Martin loves entrepreneurship and has helped dozens of entrepreneurs by validating the business idea, finding scalable customer acquisition channels, and building a data-driven organization. During his time working in investment banking, tech startups, and industry-leading companies he gained extensive knowledge in using different software tools to optimize business processes.

This insights and his love for researching SaaS products enables him to provide in-depth, fact-based software reviews to enable software buyers make better decisions.