The Analogy of AI and the Fourth Industrial Revolution

Partner Content

This author account is for Partner Content with Emerj advertising and content partnership clients. Learn more at Emerj.com/partnerships.

The Analogy of AI and the Fourth Industrial Revolution

Emerj’s own Daniel Faggella will be speaking at BootstrapLaps Applied Artificial Intelligence Conference on April 12th, 2018. Leading up to the #AAI18 event, we’ve partnered with BootstrapLabs to spark a conversation around AI in industry by putting together an article about Dan’s recent conversation with  Nicolai Wadstrom, Founder and CEO of BootstrapLabs.

The video below features a discussion between Dan and Nicolai on the impact of artificial intelligence for the industry and explore the validity of the AI’s stake as the vanguard of the fourth industrial revolution. BootstrapLabs, is a silicon valley venture capital firm, founded in 2008 and focused on investing in applied artificial intelligence applications. The discussion took place at the AI World Conference in late 2017.

More on Nicolia Wadstrom: He is a veteran entrepreneur and angel investor with over 14 years of experience in venture capitalism for technology startups. In the past, he has also served as the Director of the Board for the Swedish-American Chamber of Commerce of San Francisco/Silicon Valley (SACC-SF/SV) for a period of around 6 years. He is a frequent speaker and mentor at several American and European universities.

AAI18 banner

Conversation Highlights with Dan and Nicolai

(2.30) Daniel Faggella: In recent times, there has been an increased interest in the community about whether AI is as transformational for the human race as the earlier leaps in technology like steam or electricity; where do you stand on this?

Nicolai Wadstom: In my personal experience, I haven’t really come across any technology for the industry, that could have a greater impact than AI. Already, there have been successes in the commercialization of AI platforms which can be used to build intelligent systems that automate intelligent labor. There are a couple of underlying factors for these improved capabilities which has pushed AI technology much closer to an inflection point:

  1. As semiconductor fabrication technologies have improved vastly over the last 5 years, data processing capabilities have steadily increased in accordance with moore’s law
  2. Data storage capabilities are doubling every 12 months or so, and in addition, data transfer rates have been doubling about every 9 months.

With improvements like these, I expect AI technology to be increasingly capable in performing complex tasks, especially in areas like pattern recognition in application with massive data sets.

(9.53) DF: In the past industrial revolutions, the average workers’ typical tasks were radically transformed. In what ways will AI change the way we work today? More specifically, what types of white collar jobs do you see getting replaced in the future? Could you give us a few current use cases?

NW: In a divergence from the past industrial revolutions in the AI powered fourth industrial revolution, humans are now the ‘overseers’ of the machines. Current AI systems are really good at identifying patterns in applications where the amount of data is too large for human counterparts to effectively scour through. As a general thumb rule, we can say that most data-information-rich type of white collar jobs will be replaced very soon.

  • A few other examples along this line would be manual research jobs like paralegals or manual research in the financial industry. On the contrary, many blue-collar jobs are highly complex tasks for an AI to accomplish without human intervention.  In short, we are probably going to have plumbers for a far longer time than paralegals, because AIs are only more effective than humans in only high volume data tasks at present.
  • Another good use-case example here would be in oncology for diagnosis. A typical human oncologist would have to read about 600 pages of information and co-relate that to around 700,000 research papers for each patient; a task that is humanly impossible to put into effect more efficiently than an AI. In such an application the specific task being automated is the reading of the 600 page patient report, understanding the knowledge layer under the text and then connecting this information to clinical trials that can be applied to the patient.
  • In the venture capital sector, there are currently projects underway to understand how to use data and ML to optimize the post investment phase. This is focused on driving the right skills and talent at the right time to the startups. AI/ML platform fed with data from web scrapers would be vastly more efficient at this task than any human researcher.  The constraint here is the data, more specifically the proprietary data. If the ‘hidden’ proprietary data can be added to the equation, AIs would be much better at dynamically predicting even volatile trends.

Thus,in case of massive data sets, AI is good at processing data patterns, although, humans are still better at making distant co-relations with little data.

The product-market fit for for current AI applications in white collar services would be tailored towards augmenting human capabilities in the near future.

(21.58) DF: In the same vein, in the near future, can we expect a lot of opportunities for humans to be more productive through augmentation with AI? What is as good current example of this?

NW: I can quote an example in the psychology space: currently, around 65 million Americans undergo psychotherapy for about $150 per session on average per year. In reality though, the number of people who need psychotherapy is probably far higher.

The startup Sibly’s Mobile app provides a platform where audiences can find affordable and effective mental health coaching through conversations. AI platforms take in all the conversation data and offers suggestion to Sibly’s mental health coaches and help in consistent profile building over time.  The coaches are not licensed therapists, but trained by them, but also continuously trained by the machine.

Ultimately, the transferable use case, for such an AI system is aggregate learning and augmenting the overall skill level of a group of people.

AAI18 banner

2018 Applied Artificial Intelligence Conference

#AAI18 is taking place on April 12, 2018, and will feature presentations from artificial intelligence researchers, executives, and thinkers, including:

  • Richard Socher, Chief Scientist, Salesforce
  • Sangeeta Chakraborty, Chief Customer Officer, Ayasdi
  • Zachary Hanif, Director – Center for Machine Learning, Capital One
  • Nadeem Sheikh, VP Self Driving Programs, Lyft
  • and many more

This year’s agenda focuses on the near-term and long-term impact of AI applications across a variety of industry segments, including:

  • Transportation
  • Healthcare
  • Finance
  • Future of Work
  • Cybersecurity

 

This article was written in partnership with BootstrapLabs. For more information about content and promotional partnerships with Emerj, visit the Emerj Partnerships page.

Initial transcription and writing for this article was completed by Emerj Content Lead Raghav Bharadwaj.

Header image credit: MarTechToday

Subscribe
subscribe-image
Stay Ahead of the Machine Learning Curve

Join over 20,000 AI-focused business leaders and receive our latest AI research and trends delivered weekly.

Thanks for subscribing to the Emerj "AI Advantage" newsletter, check your email inbox for confirmation.