ML insider 2022 key results & highlights
The ML Insider is cnvrg.io’s annual survey of the Machine Learning industry highlighting key trends, points of interest, and challenges that AI developers experience. In 2021, ML Insider collected insights from 290 AI professionals who are building, training, deploying, and adapting their machine learning stack to better suit today’s modern, complex ML workflow.
In cnvrg.io’s 2nd annual ML Insider survey, we surveyed over 430 AI professionals to gain insight into the progress of AI development.
Key results and analysis overview
2022 has certainly been a tumultuous year for the global economy, especially for the tech industry. In August 2021, tech was affected by “the great resignation”, a term coined by Anthony Klotz in which over 47 million Americans voluntarily quit their jobs. Since May, corporations of every size have been affected by budget cuts, layoffs, and more as inflation and the threat of recession looms. Through all of this, the ML Insider was made to spotlight how AI is fairing through this journey in spite of it all.
Unlike many other hyped technologies, AI is faring quite well. Despite troubling times, AI technology is evolving. Overall, AI is becoming more accessible and is being adopted across industries, and company sizes, improving more business units across the organization and covering a variety of use cases. Data scientists are not solely responsible for building AI solutions. As common AI use cases become more prevalent in applications, a new definition of “AI Developer” brings engineers and software developers to the forefront of AI development.
89% of organizations are seeing the benefits of their AI solutions, and getting management buy-in is no longer a barrier. Many organizations have outgrown the challenges of the past, with the help of new tools and technologies. MLOps, automation, and monitoring tools have become more widely adopted, and are key to enabling the success of AI.
In an effort to break down silos across organizations, ML Insider aims to expose how AI is being implemented, and how challenges are shifting. The results will equip AI developers everywhere with more knowledge and transparency so that they are able to band together and influence AI adoption and maturity.
Key takeaways and trends
- More people in the organization take part in AI creation
The role of an AI developer is evolving with equal responsibility distributed between data scientists, engineers, and software developers.
- Reducing technical complexity is the key to universal AI adoption and acceptance
Technical complexity is holding AI back from achieving universal AI adoption and acceptance as a technology. Most respondents believe that technical complexity of AI development is the biggest barrier to universal adoption.
- Operationalization of AI is still heavily dependent on Developers, DevOps and Engineering
64% of respondents that found it difficult to successfully execute ML rely on Developers/DevOps/Engineering to operationalize their AI models
- 89% of organizations are seeing the benefits of their AI solutions
The majority of organizations investing in AI report that they are benefiting from their AI solutions.
More people in the organization take part in AI creation
Historically, data scientists have been solely responsible for building AI solutions and had to be highly skilled in research, engineering, and programming. The survey shows that the responsibility of AI development is evolving. As respondents indicate here, there is a changing definition of ‘AI developer’. Data scientists, engineers, and developers are equally responsible for building AI solutions in an organization.
AI is now more commonly used in applications and in more and more cases this relies heavily on the skills of engineers and developers to deploy and maintain those models in production. In addition, AI development is becoming more accessible, allowing different roles to build AI. In order to stay competitive, companies will want to cater to the challenges of those specific roles and simplify the AI process. By enabling more professionals to build AI solutions, organizations will also relieve hiring challenges.
Who is building AI solutions in your organization?
Learn more about team sizes, AI stacks, and more in the full report.
Reducing technical complexity is key to universal AI adoption and acceptance
The majority of respondents believe the technical complexity of AI development will be the biggest challenge to universal AI adoption and acceptance. Today, it is quite difficult to build AI, and there is little transparency and understanding about how these applications are being made. In order to achieve universal AI adoption and acceptance, there will need to be simpler ways to both develop and interact with AI applications.
What do you believe is the biggest challenge to universal AI adoption
Find out what other challenges AI developers are experiencing in their ML pipelines in the full report.
Operationalization of AI is still heavily dependent on Developers, DevOps and Engineering
66% of respondents rely on Developers, DevOps and Engineering to operationalize their models. The technical complexity of operationalizing AI models requires the technical skill of developers, engineers, and DevOps which can add bottlenecks and increase time to production.
I rely on Developers/DevOps/Engineering to operationalize my model
89% of organizations are seeing the benefits of their AI solutions
Your organization may be considering whether or not to invest in AI solutions. 89% of respondents reported that they are in fact seeing the benefits of their AI solutions.
Is your organization seeing the benefits of your AI solutions?
See how organizations across industries are applying AI to improve their business with the full ML Insider report.
There is no doubt that artificial intelligence in the industry is growing more valuable to an organization’s bottom line. However, with 57% of respondents reporting a low maturity, there is still a long way for AI to grow and become more and more adopted and accepted across different industries. AI maturity is still in its infancy and has a lot of potential for further development. Today, many organizations have hurdled the challenge of making AI operational, with 45.8% of respondents reporting that they are not struggling to operationalize their AI. According to ML Insider results, organizations will need to focus on making AI accessible, explainable, and less technically complex to develop.
It will be the responsibility of AI developers to set the tone for AI going forward, with 76.8% of all AI responsibilities leaning on data scientists, engineers, and developers. Leading infrastructure technologies and automation solutions will need to innovate and create tools to support all AI developers and reduce technical complexity at all phases of the AI pipeline. Enabling not only data scientists, but engineers and software developers to successfully build AI will be key to the advancement of AI, as most respondents found AI knowledge and expertise, as well as hiring AI talent to be a top challenge.
Download the full ML Insider 2022 Report for the complete results, and dive deeper into the state of machine learning.