Uploaded August 2025 | Updated September 2026, 2 days ago
#AI
This podcast challenges the common perception that Artificial Intelligence (AI) models are inherently on an unstoppable path to success and sentience, highlighting a significant rate of failure and degradation among them. It compares this phenomenon to the selective memory people have for past music, where only the most successful examples are remembered while numerous failures are forgotten. We identify several key reasons for AI model failures, including operational complexity, issues with data quality and quantity, misalignment with business needs, difficulties in production transition, and integration challenges with existing systems. Finally, we suggest potential solutions like automated data discovery and the use of Natural Language Processing (NLP) to improve data handling, emphasizing the critical need for careful planning and a contingency strategy when developing and deploying AI models.
AI was used to assist in the creation of this video.
Other podcasts in this series on YouTube are indexed here
gsfsoftware.co.uk/PBTutorials/Podcasts.htm
I don't work for, nor am I paid by, or sponsored by, the owners of PowerBASIC or any other company mentioned in this podcast, I'm just a fan of their products.
For more information on PowerBASIC visit
pbusers.org
OR
pump.richheimer.de/index.php
Source files for these projects are available from here
gsfsoftware.co.uk/PBTutorials/Projects.htm
Music by ghosthack.de
#AI
This podcast challenges the common perception that Artificial Intelligence (AI) models are inherently on an unstoppable path to success and sentience, highlighting a significant rate of failure and degradation among them. It compares this phenomenon to the selective memory people have for past music, where only the most successful examples are remembered while numerous failures are forgotten. We identify several key reasons for AI model failures, including operational complexity, issues with data quality and quantity, misalignment with business needs, difficulties in production transition, and integration challenges with existing systems. Finally, we suggest potential solutions like automated data discovery and the use of Natural Language Processing (NLP) to improve data handling, emphasizing the critical need for careful planning and a contingency strategy when developing and deploying AI models.
AI was used to assist in the creation of this video.
Other podcasts in this series on YouTube are indexed here
gsfsoftware.co.uk/PBTutorials/Podcasts.htm
I don't work for, nor am I paid by, or sponsored by, the owners of PowerBASIC or any other company mentioned in this podcast, I'm just a fan of their products.
For more information on PowerBASIC visit
pbusers.org
OR
pump.richheimer.de/index.php
Source files for these projects are available from here
gsfsoftware.co.uk/PBTutorials/Projects.htm
Music by ghosthack.de
