Click to learn more about author Ben Lorica.
Back in May, I wrote an article for DATAVERSITY® Despite Investment in Deep Learning, AI Talent Pool is Shallow that explored how companies are putting Artificial Intelligence (AI) to work through Deep Learning. The article detailed how, despite the hype and desire to integrate this powerful technology into operation, the lack of access to skilled talent was holding many organizations back. Only a few months have passed, and the rapid pace of technology evolution and adoption got me thinking: where does enterprise AI stand today?
At the time, a survey discussed in the article found that one-fifth of respondents pointed to a technology skills gap as one of the reasons why some organizations haven’t integrated Deep Learning. In the same token, 75 percent of respondents indicated their company had some combination of internal and external training programs to address this issue. Has there been progress since then or is Deep Learning still an AI pipe dream? To answer this question, we’ve looked at data to monitor interest in topics relevant to building AI products and systems, specifically areas that also warrant investment in skills development.
Here are some of the major updates regarding the state of enterprise AI:
AI Searches are Growing
Through the end of June 2018, there was double-digit growth in key topics associated with AI. Usage metrics encompass many content formats including books, videos, online training, interactive content and other material. Deep Learning and Neural Networks was the top search term (77%), with the broader AI following (72%). Deep Learning framework, PyTorch was another topic that exhibited strong growth over the last several months, indicating that professionals are getting more specific in their understanding for AI-related topics.
New Topics are Emerging
It’s one thing to learn about an individual technology or a specific class of modeling techniques, but ultimately, organizations need to be able to design robust AI applications and products. This involves hardware, software infrastructure to manage data pipelines and elegant user interfaces. As such, TensorFlow has remained the most popular deep learning framework by far, but PyTorch is also beginning to build a devoted following. Looking closely at interest in topics within Data Science and AI, Reinforcement Learning, PyTorch and Keras, an open source Neural Network library written in Python, have risen substantially this year. These newer, emerging topics of interest are a good sign for enterprise AI adoption – or at least advancing in the right direction.
Machine Learning is Here to Stay
While growth is strong across many topics associated with AI, information workers remain very interested in Machine Learning (ML), particularly in Deep Learning. Interest in ML compares favorably with other areas of technology, and alongside Kubernetes and Blockchain, it has been one of the fast-growing, high-volume search topics year-over-year.
New AI/ML Challenges Arise
We can’t talk about enterprise AI and ML without noting the importance and emphasis on data privacy, ethics and security. Users of these technologies are beginning to seek more transparency and control over their data as regulators are beginning to introduce data privacy rules and regulations. There are an emerging set of tools and best practices for incorporating fairness, transparency, privacy and security into AI systems, and fortunately, data professionals are taking note. According to data 67 percent of users – from those just exploring AI to early adopters and advanced users – explainability and transparency is already part of their model-building checklist, with compliance and privacy following close behind.
From the shifts in AI- and ML-related topics, emerging usage areas and a growing interest in ethics and privacy among data professionals, a lot has transpired in the world of enterprise AI recently. It will be interesting to see what the coming months bring and what new challenges and opportunities will arise in these thriving areas of Data Science.