Landing AI -AI transformation PlayBook - Quick Study

  • Recommendations are primarily meant for larger enterprises with a market cap from $500M to $500B.
  • First few AI projects needs to be successful rather than being most valuable.
  • AI solution can be build in 6-12 months when working with internal teams sharing the deep domain knowledge. Identifying technically feasible projects is a important along with having clearly defined and measurable objective to create business value.
  • Building an in house AI team should be considered as a long term project. AI team could sit under CTO, CIO or CDO. An Inhouse AI team would have roles like ML engineer, Data Engineer, Data Scientist and AI product Manager.
  • Providing Broad AI training - Flipped Classroom pedagogy is gaining traction especially in AI community. Employee training is now much more affordable and Learning and Development team could easily come up with study groups, MOOCs community to motivate employees to gain AI Knowledge.
    • Executives and senior business leaders: ( ≥ 4 hours training)
    • Leaders of divisions carrying out AI projects: (≥12 hours training with willingness to upskill.)
    • AI engineer trainees: (≥100 hours training with continuous learning)
  • Data is key asset for AI system, Strategic Data Acquisition is required. Unified data warehouses comes valuable for an AI system.
  • AI communication strategy is very much necessary with Investors, Regulators(government), Customers and with Talent. Clear Internal AI communication strategy can help reduce the uncertainty revolving around AI strategy.
  • AI transformation program may take 2-3 years but in 6-12 months results would start to come in and in 2 years should be able to break even depending on the quality and impact of AI system which is introduced.