The pace with which AI entered our lives is phenomenal.
It modified most individuals’s notion of synthetic intelligence. I keep in mind seeing AI as a expertise that merely delivers an output, equivalent to a suggestion or motion, based mostly on the enter. Like when a bit of software program detects a buyer’s frustration of their voice and flags them as a prime precedence.
Many firms did one thing comparable whereas advertising themselves as AI-powered within the preliminary section of AI adoption. However the actual potential got here with generative AI.
This is perhaps a shock, however lengthy earlier than ChatGPT and DALL·E grew to become family names, some enterprises had been already utilizing AI in 2017.
This text covers AI adoption statistics from 2017 to 2025. It reveals the yearly traits and the way we entered the present AI hype.
AI adoption: Evolution of clever expertise at a look
| Yr | Enterprise AI utilization | Shopper AI utilization |
| 2017 | 20% of corporations used AI | – |
| 2019 | 58% of corporations had been reported to make use of AI | – |
| 2020 | 50% of corporations used AI | 4.2 billion voice assistants in use |
| 2023 | 55% of corporations adopted AI, and 33% used generative AI | Open AI’s ChatGPT reached 100 million customers |
| 2024 | 72% of corporations had been reported to make use of AI | 8.4 billion voice assistants (projected) |
| 2025 | 92% of firms plan to put money into Gen AI over the subsequent three years | Elon Musk claimed that AI will grow to be “vastly smarter” than people in 2025. What are you anticipating to see this yr? |
Sources: McKinsey: State of AI in 2024, McKinsey: AI within the Office, Statista, Reuters, and NY Instances
AI adoption from 2017 to 2025: 9 years in overview
Beneath are a couple of statistics that showcase how AI has developed over the previous 9 years.
2017: 20% of enterprises adopted AI
In 2017, solely 20% of survey respondents confirmed they adopted AI in at the least one enterprise space.
McKinsey’s State of AI in 2022 lined AI evolution between 2017 and 2022. The survey noticed that although AI adoption was 2.5x larger in 2022 than in 2017, it elevated to 58% in 2019 however dropped to 50% in 2022.
In addition to development, listed here are a couple of further occasions that occurred round AI adoption in 2017:
- The variety of AI papers revealed annually elevated 9 instances in comparison with the 1996 information.
- The variety of lively US startups growing AI techniques elevated 14 instances since 2000. In 2017, round 600 startups developed AI techniques. The annual VC funding elevated by 6x in the identical interval.
Six instances
improve in AI vibrancy was noticed in 2017 in comparison with 2000 information. AI vibrancy is the measure of the liveliness of AI as a area.
Supply: Stanford
- Job openings that wanted AI expertise within the US elevated by 4.5x since 2013.
- Error charges in picture labeling fell from 28.5% in 2010 to 2.5% in 2017.
- An AI system skilled on a dataset of 129,450 medical pictures of two,032 ailments might classify pores and skin most cancers at a degree of competence similar to that of a dermatologist.
- In 2017, the proportion of company AI papers within the U.S. was 6.6x that of company AI papers in China.
2018: AI Adoption grew by greater than 100%.
The McKinsey report estimated that AI adoption was 47% in 2018, in comparison with solely 20% in 2017. That is greater than double the adoption charge in 2017, exhibiting greater than a 100% improve in adoption.
On the tutorial aspect, AI was a trending subject. The 2018 Development of Synthetic Intelligence (AAAI) convention was held in February in New Orleans, Louisiana. The convention noticed that 70% of papers submitted had been affiliated with the usor China. Nevertheless, the variety of papers accepted was remarkably even, 268 and 265, respectively.
Beneath are some extra fascinating occasions that occurred round AI adoption in 2018.
- The U.S.-affiliated papers had an acceptance charge of 29% vs. China-affiliated analysis papers at 21%.
- Attendance on the Worldwide Convention on Studying Representations (ICLR) 2018 grew 20x since 2012. The convention targeted on deep reinforcement studying inside AI.
71%
of the applicant pool for AI jobs within the U.S. was males. Workshops like girls in machine studying (WiML) and AI4All inspired participation from different underrepresented teams.
Supply: Stanford
- WiML alone noticed 750+ girls take part — a 600% improve from 2014.
- Energetic AI startups within the U.S. elevated 113% between 2015 and 2018.
- Articles on AI grew to become 1.5x extra optimistic from 2016 to 2018
Based mostly on these statistics, AI adoption trended upward in 2018. Corporations like Amazon and Alphabet invested $16.1B and $13.9B in analysis and improvement associated to AI, respectively.
2019: AI adoption in organizations reached 58%
Contemplating the continued improvement and analysis in AI, 2019 confirmed a rise in AI’s adoption charge. In addition to this, many occasions occurred within the AI house throughout this era, together with:
- The share of AI-related job postings within the U.S. rose from 0.26% in 2010 to 1.32% by October 2019. Machine studying topped the charts and made up 0.51% of all job listings.
- Washington had the very best share of AI job postings at 1.4%, adopted by California and Massachusetts at 1.3%, New York at 1.2%, DC at 1.1%, and Virginia at 1%.
58%
of huge firms used AI in at the least one space in 2019. Nevertheless, solely 19% addressed algorithm explainability dangers, and a mere 13% labored to cut back AI bias and promote equity.
Supply: Stanford
- In 2019, non-public international funding in AI crossed $70 billion.
- Startups raised over $37 billion, and mergers and acquisitions (M&A) had been valued at $34 billion. IPOs introduced in $5 billion, and minority stakes attracted round $2 billion.
From mid-2018 to mid-2019, over 3,600 international information articles explored AI-related matters. They had been primarily targeted on equity, interpretability, and explainability.
2020: AI adoption in massive enterprises fell to 50%
COVID-19 clearly confirmed its influence on AI adoption in 2020. With priorities shifting and budgets tightening, the adoption charge fell to 50% in 2020. Nonetheless, not every little thing slowed down.
- Funding in AI for drug design and most cancers analysis hit $13.8 billion in 2020, 4.5x larger than in 2019. This was the top-funded AI sector globally.
- In 2020, applied sciences like facial recognition, video analytics, and voice ID grew to become extra correct, reasonably priced, and widespread. It drove broader surveillance capabilities worldwide.
- Though many teams produced moral frameworks and rules, a constant technique to measure or consider AI improvement was absent.
- In 2020, round 4.2 billion digital voice assistants had been predicted for use worldwide.
- The U.S. posted 8.2% fewer AI jobs in 2020 than in 2019. The roles dropped from 325,724 to 300,999.
Though general adoption was comparatively slower, some sectors, like healthcare, noticed excessive funding in AI-related R&D.
2021: AI adoption climbed again to 56% in organizations
Some survey respondents recommended their AI investments didn’t improve regardless of the worldwide COVID-19 pandemic. Participation in in-person occasions moved to an internet medium as an alternative. Right here’s a snapshot of some fascinating occasions within the AI house in 2021.
- AI journal publications grew by 34.5% from 2019 to the beginning of 2021.
- Main AI conferences shifted on-line attributable to COVID-19. Consequently, the attendance doubled throughout 9 main occasions.
- Generative AI made substantial progress. AI was capable of create lifelike textual content, audio, and pictures.
25%
of survey respondents reported that at the least 5% of their organizations’ earnings earlier than curiosity and taxes (EBIT) had been attributable to AI in 2021.
Supply: McKinsey
- AI got here nearer to human efficiency in language duties. On fundamental studying benchmarks (e.g., SuperGLUE), AI was capable of beat people by 1–5%. For complicated duties like abductive pure language inference (aNLI), the human-AI efficiency hole shrank from 9 factors in 2019 to only one in 2021.
- Robotic arms grew to become extra reasonably priced. The median value dropped by 46.2% in 5 years, from $42,000 in 2017 to $22,600 in 2021. This made robotics analysis extra accessible.
- AI funding surged to $93.5B in 2021, greater than double 2020 ranges.
Generative AI rose in 2021, and lots of large organizations elevated their investments. Personal funding in AI reached roughly $93.5 billion in 2021, greater than twice the quantity invested in 2020. Nevertheless, the variety of newly funded AI startups declined from 1,051 in 2019 to 762 in 2020 and additional right down to 746 in 2021.
2022: AI adoption plateaued at 50%
After bouncing again, enterprise AI adoption dipped once more to 50% in 2022, indicating that the {industry} had reached a plateau. Nevertheless, this development was just like different applied sciences within the early years of their adoption.
Michael Chui, Accomplice at McKinsey International Institute, mentioned, “We is perhaps seeing the truth sinking in at some organizations of the extent of organizational change it takes to embed this expertise efficiently.”
Corporations that thought implementing AI can be a fast train had been discouraged, however those who grew their AI muscle slowly included extra AI capabilities.
Right here’s an summary of every little thing that influenced AI adoption ultimately in 2022:
- The U.S.–China collaboration in AI analysis noticed essentially the most cross-country exercise from 2010 to 2021. It elevated by 5x over the last decade regardless of rising geopolitical tensions.
- Corporations used extra AI instruments. On common, 3.8 totally different AI capabilities in 2022, up from 1.9 in 2018.
- The most typical use of AI remained the identical for 4 years: firms leveraged AI to optimize service operations.
40%
of firms utilizing AI spent greater than 5% of their digital funds 5 years in the past. In 2022, over half of them spent that a lot or extra.
Supply: McKinsey
- 54% of the world’s massive language and multimodal fashions in 2022 got here from U.S. establishments.
- In 2022, the {industry} dominated AI improvement. Trade produced 32 main fashions, in comparison with simply three from academia.
- The yr 2022 noticed the general public launch of generative instruments like DALL·E 2, Secure Diffusion, ChatGPT, and Make-A-Video.
- For the primary time in a decade, international AI non-public funding declined 26.7%, from $93.5B in 2021 to $91.9B in 2022. Nonetheless, AI funding in 2022 was 18x larger than in 2013.
- 37 international locations handed AI legal guidelines in 2022. This quantity elevated from 1 firm in 2016.
- The U.S. remained the worldwide chief in AI funding, and attracted $47.4B in 2022, 3.5 instances greater than China.
Though funding decreased and adoption plateaued, AI fashions, legal guidelines, insurance policies, and capabilities rose in 2022. This was seemingly when the group gained maturity in implementing AI, main the {industry} towards the AI spring.
2023: AI adoption grew to 55%
In 2023, AI adoption in organizations elevated to 55%, whereas 33% of survey respondents confirmed that their agency used generative AI ultimately.
- In 2023, industry-led AI analysis launched 51 key machine studying fashions, whereas academia contributed solely 15. Collaborations between {industry} and academia hit a document with 21 joint fashions.
- AI organizations launched 149 basis fashions in 2023 — greater than double the quantity in 2022. Almost 66% of those fashions had been open-source, up from 44.4% in 2022.
- U.S.-based establishments produced 61 prime AI fashions in 2023, greater than the EU (21) and China (15) mixed.
- China led the world in AI patent origin with 61.1% in 2022. The U.S. adopted with 20.9%, down from 54.1% in 2010.
- Generative AI funding skyrocketed to $25.2B in 2023, 8x greater than in 2022.
67%
of firms had been anticipated to extend their AI investments over the subsequent three years.
Supply: McKinsey
- The U.S. AI funding hit $67.2B in 2023, practically 9 instances greater than China. China and EU investments fell sharply.
- The U.S. AI laws grew by 25 in 2023, up from only one in 2016.
- 21 U.S. businesses regulated AI in 2023, up from 17 in 2022.
Investments in generative AI elevated in 2023, creating the preliminary setup that fueled its rise in 2024 and 2025. Let’s take a look at what precisely occurred in 2024.
2024: 72% of organizations used AI
From 50% adoption in 2022, AI adoption surged to between 72% and 78% in 2024, relying on which examine you belief extra.
Personally, I really feel the adoption charge is a little bit larger than what’s been reported. Stanford’s 2025 AI Index reported, “78% of organizations used AI in 2024.” Both manner, it’s clear that the AI adoption trended upward. McKinsey and Stanford’s information mirrored this.
Listed here are a couple of related statistics to this adoption development:
- In 2024, the {industry} constructed nearly 90% of the highest AI fashions, up from 60% in 2023. Academia led in essentially the most extremely cited AI papers.
- The U.S. produced 40 prime AI fashions in 2024, forward of China (15) and Europe (3)
- Video era from textual content improved in 2024. New instruments like OpenAI’s SORA and DeepMind’s Veo 2 create higher video content material.
- AI incident reviews hit a document with 233 circumstances in 2024, up 56% from 2023.
- 8 out of 10 professionals prioritized AI capabilities when deciding on software program.
- 40% of firms relied on automation to streamline and enhance information entry.
- 83% of organizations that bought an AI answer noticed a optimistic ROI.
- 75% of companies applied two to 5 AI options, suggesting a measured but dedicated method. In the meantime, 17% have adopted six to eight options throughout their operations.
- Solely 2% of organizations reported fast AI adoption in IT. Nevertheless, advertising emerged because the quickest division to undertake AI, in accordance ot 53% of organizations.
- Solely 26% of firms turned AI pilots into actual enterprise worth. Furthermore, solely 4% had been on the slicing fringe of AI maturity.
- 62% of AI worth got here from core enterprise areas like operations, gross sales, and R&D.
- Solely 10% of AI implementation challenges got here from AI algorithms, but many firms wrongly overfocused right here. 70% of obstacles had been people- and process-related.
- BCG reported that with out decisive motion, 75% of firms risked falling behind within the AI race.
- Mentions of AI in international legislative data rose 21% in 2024.
- The U.S. federal businesses launched 59 AI laws in 2024, up from 25 in 2023.
- AI optimism rose globally, from 52% in 2022 to 55% in 2024.
- 60% of individuals believed AI would change their job; solely 36% feared it will substitute them.
AI adoption reached an all-time excessive, with a charge between 72% and 78% globally. With continued funding, this charge is predicted to rise even additional in 2025.
2025: Getting into the clever age
McKinsey analysis suggests that the majority firms are investing in AI, however just one% imagine they’re at maturity. The problem shouldn’t be workers however leaders who should not transferring quick. Whereas firms are wanting ahead to the long-term features of AI, 92% are planning to extend their funding within the subsequent three years.
Beneath is an summary of what’s newest and but to come back within the AI house in 2025.
- 69% of C-suite firms started investing in generative AI over a yr in the past. Regardless of that, solely 47% say they’re making sluggish progress in constructing Gen AI instruments.
- C-suite leaders are 2.4x extra prone to say worker readiness is a big barrier to adopting AI. However workers are utilizing generative AI 3 times greater than their leaders assume.
- 48% of workers rank coaching as essentially the most essential issue for gen AI adoption.
These stats and different ongoing traits recommend that AI isn’t plateauing anymore; it’s on the rise from an innovation, implementation, and adoption perspective.
An MIT Know-how Assessment article suggests that giant language fashions (LLMs) will be capable to “motive.” It talks about 2023 being the age of generative pictures and predicts 2025 to be the age of generative digital playgrounds.
Yr-by-year development traits in AI adoption
To focus on the traits, here’s a transient year-by-year abstract of notable AI adoption milestones:
- AI was principally in R&D (2010 to 2016). Early client merchandise like Apple’s Siri (2011) and Amazon Alexa (2014) launched AI to the general public, however enterprise use was uncommon. No exact international adoption surveys exist for this era.
- Surveys started (2017): A landmark McKinsey examine in 2017 discovered 20% of firms deployed AI in some perform. The Trade buzz intensified, however broad adoption was nonetheless rising.
- Speedy development (2018–2019): Roughly half of the corporations had experimented with AI. Generative AI was not but mainstream, however machine studying and automation instruments grew to become widespread in tech-focused firms.
- Consolidation and COVID influence (2020): International surveys round 2020 confirmed AI utilization round 50% in enterprises. In the meantime, client AI surged with voice assistants. The COVID-19 pandemic pushed many firms to put money into automation and distant providers. Healthcare and retail noticed AI funding jumps, and digital assistants grew to become extra widespread at residence.
- Continued adoption (2021): By 2021, many organizations had AI pilots or deployments, however adoption plateaued. Huge tech releases equivalent to GPT-3, DALL·E, and so forth. expanded AI capabilities. Nevertheless, the total influence was nonetheless in pilots.
- Breakout of generative AI (2022): ChatGPT and different generative fashions had been launched in late 2022. They captured worldwide consideration. Corporations examined generative AI for content material, code, and design duties. Shopper consciousness elevated: roughly half of the folks had heard of ChatGPT by year-end.
- File development (2023): ChatGPT reached 100 million month-to-month customers by January 2023, the quickest adoption of any client web app. On the similar time, enterprises resumed sooner AI adoption.
- Mainstream and mass deployment (2024–2025): The Stanford AI Index (2025) reviews that 78% of organizations will use AI in 2024. By 2025, main economies will improve funding in AI improvement and regulation, and {industry} surveys count on continued development.
Sector snapshots: Key AI adoption figures throughout industries
Beneath are concise stats of AI adoption by {industry}, illustrating how AI use differs throughout industries:
| Trade/sector | AI adoption development | Use circumstances |
| Finance | 72% of finance leaders use AI. | Fraud detection and threat administration |
| Healthcare | 90% of hospitals are estimated to make use of AI. | Diagnostic and monitoring |
| Retail | 53% of huge retail chains use AI | In-store analytics, demand forecasting, and buyer personalization |
| Manufacturing | 35% of producers used AI in 2023 | Predictive upkeep and high quality management |
Transferring towards effectivity
From roughly one-in-five firms utilizing AI in 2017, AI adoption grew to roughly three-in-four firms by 2024. The expansion is placing, and it is nonetheless anticipated to rise for the remaining half of 2025. AI is right here to remain and can seemingly drive folks, processes, and expertise towards higher effectivity within the foreseeable future.
As a person, it’s advisable to consider what components of your workflows will be automated by AI. If it’s saving you extra time, automate it. This may assist you to keep aggressive, efficient, and environment friendly within the altering instances. I personally don’t imagine AI will substitute people of their jobs. Nevertheless, I strongly really feel {that a} human with an AI sidekick will make the duo irreplaceable.
As an organization, implement AI strategically and construct it over time slightly than sprinting towards it. This may assist you to attain AI maturity whereas successfully realizing enterprise advantages.
Need to study extra about AI adoption? Try my colleagues’ article on AI Adoption in 2025.