Cyborg AI Minds are a true concept-based artificial intelligence with natural language understanding, simple at first and lacking robot embodiment, and expandable all the way to human-level intelligence and beyond. Privacy policy: Third parties advertising here may place and read cookies on your browser; and may use web beacons to collect information as a result of ads displayed here.

Friday, December 08, 2017


Cleaning up the interface between the thinking AI and the human user.

In we are trying to clean up the display of the thinking of the AI as output in the man-machine dialog. Code in the module EnThink() which has been displaying intermediate output may now be commented out.

At the end of Sensorium() we change the point-of-view $pov flag back to one ("1") and we stop getting a duplicate line-display during the AudInput() module, where $pov is tested for a value of two ("2") during external input and not during the re-entry of thought.

All these minor adjustments to the AGI codebase are due to a healthy respect for the principle of "survival of the fittest" in these early stages of AI evolution. We are eager for so-called "early adopters" to install the Perlmind AI on numerous computers where the surviving codebase may mutate and evolve along multiple not blood-lines but codelines. We encourage the emerging cohort of AI Mind Maintainers to post news and ideas in diverse forums devoted to Perl, to AI, to robotics, to neuroscience and so forth. Anyone who finds a bug in the AI should please report it in the Perl subReddit or in comp.lang.perl.misc on Usenet.