Draft for AI risk is corporations

src/content/blog/drafts/ai-is-not-the-threat-corporations-are.md

@@ -0,0 +1,31 @@
+---
+title: Humanity faces an existential threat
+description: Humanity faces an existential threat right now.  A lot has been made of predictions that there is a 10% chance that humanity could be wiped out in 10 years by AI.  I'm not worried.  We have been doing a pretty good job of annihilating and immiserating ourselves without AI.
+---
+
+- There are some predictions about AI threats
+  - See the 10% in 10 years tweet.
+  - The mechanism isn't all that clear.
+- This is not a technology problem, it's a human problem.
+- Corporations are like a slow AI.
+  - The key is that they are a system complex enough to operate on it's own without any human actually overseeing it.
+    - Executives claim no knowledge, but they encourage workers to push boundaries. They assume they have the blessing of their board and their job depends on hitting those targets.
+    - Directors pursue aggressive targets but expect to stop short of harm and law breaking. They assume they have the blessing of leadership and their job depends on hitting the targets.
+    - ICs cut corners and break laws assuming they will have the blessing of leadership and their job depends on it.
+    - The board encourages the executives to make things happen and focuses on the metrics and targets.
+  - Corporations are perfectly capable of causing harm to the world already.
+    - Tobacco
+    - Oil
+    - Chemical companies
+    - Whaling
+    - Factory farming
+- The actual threat of AI is in how we use it, not what we do.
+  - Someone has to give it resources
+  - Someone has to turn it on
+  - Someone has to look the other way when it starts misbehaving
+  - Someone has to ignore the warning signs
+  - AI won't kill humans, corporations will.
+- Who can stop it?
+  - People can demand more
+  - Governments can put in regulation
+  - Governments can break up the monopolies.