Artificial intelligence
fromMedium
2 days agoWhy Your AI System Is Open-Loop
Open-loop AI systems audit spending after the fact, while closed-loop systems proactively control costs through continuous measurement and adjustment.
The problem was not the agents. Every individual agent performed well within its domain. The problem was the missing coordination infrastructure between them, what I now call the 'Event Spine' that enables agents to work as a system rather than a collection of individuals competing for the same resources.
The researchers developed round fluffy robots with motorized ribcages that can simulate breathing by expanding and contracting. More than 100 participants held these robots, which breathed in a stable pattern, in an accelerated fearful manner, or not at all, while the participants watched a scary clip from The Shining. The team found that the heart rates of people holding hyperventilating robots increased the most, compared with those holding chilled-out or stationary robots.
Time pressure, limited information, confusion, fatigue, and mortality salience combine to set the stage for decision-making errors, sometimes with grave consequences. An example is the downing of Iran Air Flight 655 by a missile launched by the USS Vincennes in 1988, resulting in the death of 290 passengers and crew. In a time of heightened tension between the U.S. and Iran, the captain of the Vincennes misidentified the airliner as an incoming hostile aircraft and ordered his crew to shoot it down.
I'll be talking about holistic engineering or the practice of factoring in your technical decisions, designs, strategies, all the non-technical factors that are actually forces that influence your organic socio-technical problem space. As much as you can see in this canyon how natural forces have influenced the shape of the earth, so you can see the color. You can see all the different layers.
After years of computer saying no, and giving us all migraines and premature grey hair, I'm starting to worry that computer or rather AI large language models like ChatGPT and Gemini are taking too much of a fancy to playing nice and saying yes. I confess to using both of these programs, but I've noticed that, well, it's as if they're trying to please, with statements like You're absolutely right, Jeff, and That's pretty much right.
Harry frowned. "I'm not seeing the value in it. Can you explain it clearly? Is there any other solution?" Tom leaned in. "This isn't making much sense. You could try this instead. It's simpler." Leina sighed. "Next time you present, put more thought into your reasoning." Meanwhile, Ron trembled with anxiety. He wanted to make a point but ended up rambling. This was his second failed attempt at defending his ideas.