Time flies when you're having fun. 😀 Five years already since I first disclosed some of the detail on the KOS and SGM to the public -- initially at NM's leading IT conference and then several additional. I decided to have the first one recorded and share.
Although I'm proud of our work and personally enjoy sharing and discussing, from the founder and entrepreneur's perspective, I wouldn't advise it. Our IP attorney certain wouldn't. A few reasons why:
We like to think that customers are trustworthy and wise enough to adopt better systems when they become available, but that's often not the case. Many things influence adoption. Wisdom isn't near the top. Protectionism usually is.
Our R&D disclosures with the most traffic produced the worst results (i.e. our semantic enterprise paper on our healthcare platform in 2010. 10 million + reads, most healthcare orgs in the world, yet zero responses from prospective healthcare customers).
The truth is we are primarily educating potential competitors -- internal in organizations as well as external like Big Techs and VC-backed ventures with obscene amounts of capital to burn.
R&D disclosures help mostly academics seeking to advance their careers, and in particular AI scientists attempting to find a lucrative job in Big Techs or over-funded start-ups. It's usually due to other academics hiring them -- the guild at play.
The greatest value for entrepreneurs in disclosing R&D is if they are just seeking to flip their ventures, whether a bolt-on technology or an acquihire as it does reveal value and what the individuals are working on.
Also need to be careful with governments. They don't invest $10s of billions annually in espionage for no reason. Governments will usually glean as much as possible only to hire their favorite contractors (typically full of retired gov't workers) to build a copycat.
Whether government, corporate, or over-funded VC-backed firms (mostly in SV), that they usually fail is unsurprising as anyone with much intelligence will withhold critical elements of technology to make it very difficult to reverse engineer, but it doesn't stop many from trying. It's interesting but sad.
What does work for entrepreneurs? Careful selection of targets, private demonstrations that protect IP, and strong encryption. Pre-existing trusted relationships are always best, which only serves to highlight the toxicity for inventions in this environment.
Also be careful with NDAs as even those with good IP protection typically have short terms with expiration dates, and in most cases large orgs require years to commercialize anyway.
Bottom line: It's a minefield out there. The Web and social networks are full of them (mines).
It's been five years this month since Mark did his first talk on Enterprise AI and the KOS at the ExperienceIT NM conference. This is the leading IT conference in New Mexico. The audience was primarily IT professionals from national labs in NM (LANL, Sandia, AFRL).
Although much has taken place under the banner of AI and enterprise AI over the last five years, most of the issues of importance to business leaders haven't changed, particularly relating to governance, security, productivity and human behavior.
This was the first public disclosure on the KYield OS (KOS) and the Synthetic Genius Machine (SGM), representing 23 years of R&D at the time. Mark has since provided similar talks to IEEE Rising Stars, university systems, and major corporations.
https://lnkd.in/gecbqvXA
