US AI Standards Document

August 15, 2019

DarkCyber learned about “U.S. LEADERSHIP IN AI: A Plan for Federal Engagement in Developing Technical Standards and Related Tools.” The research team poked about and located the report. You can, as of August 15, 2019, download the 52 page document at this link. There are some interesting “targets” in the report; for example, the need for documented case examples. A related document has been prepared by AI insiders. You can download this 112 page report at this link. A bit of Binging and Googling (coupled with patience) reveals similar position papers.

DarkCyber’s view is:

  • The authors of these documents are not upfront about the technical balance tipping from the US to other countries’ efforts
  • The spectrum of “flavors” of artificial intelligence
  • The lag between what specific companies are doing and what the bold vision of a smart future will deliver.

How can these issues be addressed? DarkCyber has been noodling about the gaps, the spectrum, and the Borges like reality and fantasy dichotomy for years.

Based on the information presented in these documents, other issues are of greater concern to those wrestling with AI in a decidedly American venue, with US athletes, and a US definition of “world champion.”

In short, a question: What if Jorge Luis Burges’s observation is correct:

Reality is not always probable, or likely.

DarkCyber assumes one could ask IBM Watson or Amazon SageMaker.

Stephen E Arnold, August 15, 2019

Comments

Got something to say?





  • Archives

  • Recent Posts

  • Meta