The survey was conducted by Atomik Research and included 2,037 professionals from 10 countries. It found that:
- 42% of respondents had experienced an AI project failure in the past two years.
- The average failure rate for AI projects was 36%.
- Despite the high failure rate, 78% of respondents believe AI can improve their organization’s capabilities or services.
- 54% of respondents said they believe AI has the potential to lead to long-term breakthroughs.
Challenges that Impact Project Success Rates
The survey also found that three main areas of friction hinder the success of AI projects: organizational, technological, and financial.
Organizational friction
Organizational friction can include a lack of data science talent, low AI literacy among the workforce, a lack of understanding of how AI can be used to improve business outcomes, and difficulty in getting different departments to collaborate on AI projects.
- 75% of respondents acknowledged the scarcity of data science talent.
- 35% highlighted a need for more AI literacy among their workforce.
- 58% of respondents identified the shortage of talent and the time required to upskill existing employees as the most significant obstacle in their AI adoption strategy.
Technological friction
Technological friction is caused by legacy systems not designed to support AI initiatives, data quality issues, and slow data processing speeds. Hence, it is difficult to make informed decisions with data. Over half of the survey participants reported encountering technical limitations that impede the progress of their data and AI initiatives, representing another significant source of friction. Respondents cited data processing speed, difficulty making quick, informed decisions, and data quality concerns among the identified technological challenges.
- 63% of respondents claimed that their organizations tend to make working with AI-driven data tools more complicated than necessary.
- 33% highlighted legacy systems’ inability to support advanced AI and machine learning initiatives as a recurring technological hurdle.
Financial friction
Financial friction is caused by the high cost of implementing AI projects and the perceived risk of failure. It can include a need for more funding for AI projects and a focus on the upfront costs of AI projects rather than the long-term benefits
Despite wanting to scale data and AI strategies, financial constraints pose a significant obstacle for many organizations.
- 25% of respondents identified financial limitations as a friction point negatively affecting AI initiatives within their organizations.
- 28% mentioned that leadership focuses too heavily on upfront costs and needs to grasp the long-term benefits of investing in AI and machine learning.
- 33% of respondents pointed to the perceived or actual high cost of implementation as a shortfall when relying on AI tools to complete projects.
Project Failure Rates
The survey results indicated that project failures are common, with one in four respondents reporting that over 50% of their projects fail. Additionally, 42% of participants admitted experiencing AI project failures within the past two years, with an average failure rate of 36% for those organizations.
Despite the high rate of project failures, organizations are optimistic about the future of AI. 78% of respondents believe that AI has the potential to level up their capabilities or services in the long run, and 54% see potential for long-term breakthroughs based on minor successes.
Organizations must address the three main areas of friction to succeed with AI. They need to invest in data science talent, upskilling the workforce, improving data quality, invest in technology that can support AI initiatives, and aligning different departments within the organization. They also need to choose the right AI tools and implement them cost-effectively and scalable.
“Organizations today recognize the imperative of using their data as a strategic asset to create competitive advantages,” said James R. Scapa, founder and chief executive officer of Altair.”
“But friction points clearly exist around people, technology, and investment, preventing organizations from gaining the data-driven insights needed to deliver results. To achieve what we call ‘Frictionless AI,’ businesses must make the shift to self-service data analytics tools that empower non-technical users to work easily and cost-effectively across complex technology systems and avoid the friction inhibiting them from moving forward.”
Challenges Across Regions
Altair’s survey highlighted regional differences globally when deploying organizational data and AI strategies. Respondents from the Asia-Pacific (APAC) and Europe-Middle East (EMEA) regions reported higher rates of AI failure in the last two years compared to the North-South America (AMER) region.
- 65% of APAC and 61% of EMEA respondents agreed that their organizations make working with AI tools more complicated than necessary.
- 78% of APAC and 75% of EMEA respondents needed more data science talent.
Overall, the survey findings suggest that AI is a promising technology with the potential to revolutionize many industries. However, organizations need to be aware of the challenges in implementing AI and take steps to address them if they want to succeed.
The Path to Frictionless AI
Altair’s survey emphasizes the need to achieve “Frictionless AI” for successful data analytics projects. This state of frictionlessness entails seamless integration of data analytics into business processes, quick and repeatable project execution, and scalability. It involves eliminating technical friction between organizations and their data, bridging the gap between data experts and domain experts, ensuring smooth workflow between data application design and production deployment, and mitigating migration friction when infrastructure or tools change.
Altair’s global survey on data and AI project implementation revealed the prevalent challenges of organizational, technological, and financial friction. The need for more data science talent, technical limitations, and financial constraints hinder project success rates. Despite these obstacles, organizations remain optimistic about the long-term benefits of AI adoption. To achieve “Frictionless AI,” businesses must embrace self-service data analytics tools, invest in upskilling their workforce, simplify AI-driven data tools, and balance investment considerations. By addressing these friction points, organizations can unlock the true potential of data and AI, driving innovation and competitive advantage.
Altair is a global leader in computational science and artificial intelligence (AI). The company provides software and services that help organizations solve complex problems in various industries, including aerospace, defense, automotive, energy, and manufacturing. Altair is headquartered in Troy, Michigan, and has offices in over 20 countries.
