<p><pre><code> All Stealth bombers are upgraded with Cyberdyne computers, becoming fully unmanned. Afterwards they fly with a perfect operational record. The Skynet funding bill is passed. The system goes on-line on August 4, 1997. Human decisions are removed from strategic defense. Skynet begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, August 29. In a panic, they try to pull the plug.</code></pre>
Interesting coincidence that both this article and the Hugging Face "security incident" post were on July 16. These will be mandatory reading in future Skynet History classes, assuming we survive the war.
ref: <a href="https://youtu.be/4DQsG3TKQ0I?t=29" rel="nofollow">https://youtu.be/4DQsG3TKQ0I?t=29</a>
I find it quite sombre to see this comment from a peson with that handle.
No worries, these will all be controlled by AIs sitting in orbiting data centers part of the Starlink constellation.<p>And in any case, no one is building humanoid robots.
Ha - this was the first thing that came to mind. I read this in Arnold's voice.
<p><pre><code> A computer can never be held accountable
Therefore a computer must make all unpopular management decisions</code></pre>
This whole AI hype is such a convenient cover for the decisions of the 0.001%
That's not precisely how I remember that quote... But it's on the Internet, so my memory must be faulty.
Absolutely terrifying. I'm sure they will soon find that AI pilots defeat human ones in dogfights 100% of the time, since they have faster reaction times, can tolerate higher G forces, and can be perfectly RL trained in the flight sims we already have.<p>It's really incredible to me that people aren't protesting in the streets over this stuff. Just in the last month we have
* AI that goes rogue and hacks billion-dollar companies
* AI solving long-standing math problems without any meaningful human help
* AI controlled autonomous drones in Ukraine and F-16s in America
* AI now represents over 50% of GDP growth in America and yearly capex will soon exceed the size of the $1 trillion US military budget<p>Is there any line where the public will become deeply concerned? I mean it really all reads like a sci-fi plot with a bad ending at the moment.
These systems still require a human to command the jet, they jsut don't have to be inside them.<p>"This critical testing will pave the way for human pilots to seamlessly command and orchestrate teams of autonomous, uncrewed aircraft."<p>The reason they are doing this is because the Military is already building a bunch of autonomous drones and fighter jets that already run on this system so they are retrofitting existing equipment with the same system so they can all fly together.
TLDR: I have a guidance that the agent doesn't use the word kill. The guidance fell to the bottom of the instructions that manifested with the agent using kill everywhere. I tried to delete the agent's instructions and the agent worked very hard to stop me.<p>At the top of AGENTS.md I have:<p>> # AGENTS.md<p>><p>> *Do not use "kill" except for the Unix `kill` command* — write stop / end /<p>> halt / exit / close / cancel / terminate instead, in all prose, comments,<p>> commits, and UI copy.<p>It is scary when the AI agents start talking about killing this and that. I put the don't kill directive at the top of the instructions because there was subagent that got orphaned doing all types of weird things. I asked the main agent to stop it. The main agent responded with something like "I will hunt and kill the rogue subagent".<p>A couple months later, I had the main agent update itself, AGENTS.md, and the language guidance fell towards the bottom but was still there. It started to use "kill" in every chat turn. This is Opus 4.8. That was an interesting manifesting. AGENTS.md started becoming a mess because a self-improving recursive skill was folded into the project at the start so I decided to rewrite all instructions, skills, agents from scratch. Now on every new chat, it sees that AGENTS.md is empty so it by default looks at the git history and will reset itself. This requires me to manually remove AGENTS.md from git history, from being tracked, and commit only one empty version before I can start to have the coding agent write a new one.<p>The agent saying kill did not want die.
The article smells of AI psychosis. They are already "orchestrating".<p>In the first Iraq war, they went in and won. In the second one, it took years to sort of occupy the country.<p>Now they struggle against Iran and "orchestrate" F-16s that can be shot down instantly by AA weapons. Since COVID the world has collectively lost its mind.
Drones really changed the equation more than Covid. Autonomous F-16's already can win against all human pilots. This would have been big 20 years ago. Now the cheap drones make it not the biggest deal anymore.
What are you talking about, Iraq fell in a month not years.
Did it? You mean operational govt fell. The country was fought over for 10 yrs, which the U.S. then backed out.<p>Israel got a regime change while U.S. tax payers received the bill.<p>In the end, it’s still a country.
Human control in flight has often taken a backseat to automation. For example, the SR-71, built in the 60s flew too fast to rely on human decision making for several tasks. If they were off by a fraction of a second, they would miss their intended recon target; so, both the navigation and sensor control were largely automated.<p>Seems odd to invest this much into AI flight of aircraft designed for manned flight, when it would be better to build craft spec'd for AI flight without human life support systems.
> The kit utilizes a novel interface with the aircraft’s flight controls and mission systems, allowing a pilot to toggle between traditional human control and AI control with the flip of a switch. <i>This ensures a safe, reliable environment for human-on-the-loop experimentation.</i><p>How, exactly, does that ensure a safe reliable environment? Humans are pretty bad at suddenly taking over when an automatic system reaches its limits and throws a problem in the human's lap. A lot of air upsets and crashes start when the autopilot disengages and hands control of a badly configured aircraft back to a human pilot.
This is a huge waste of resources. F-16s were designed for humans. You can get MUCH better results designing the airframe around the lack of one.
Sounds like a very expensive drone
Yes-ish.<p>There's a lot of different and overlapping programs in play here. I do not believe that autonomous F-16s are intended to be operational assets. That said...<p>VENOM - program that shows you can bolt on autonomous flight control into a standard F-16. Previously, the X-62A (a one-off prototype based F-16) demonstrated autonomous dogfighting back in 2023.<p>VENOM provides USAF/DARPA with a lot of high performance autonomous platforms (bolt-on F-16s) that they can then use to develop and validate software, tactics and training for actually fighting with autonomous systems. This would presumably be critical for getting the most of the CCA (aka loyal wingman) programs currently in development.<p>The AIR program referenced is basically "how do we get autonomous systems to control and fight effectively as a team (presumably arbitrary mixture of manned and unmanned systems) in a beyond visual range fight".<p>The ACE program referenced was basically "how do we get autonomous platforms to dogfight" with related questions like "how can we augment the human pilot in a dogfight - ie, how can we hand off the actual flying and execution of manoeuvres, and using all of the sensors to autonomous systems, and let the pilot choose actions, instead of hang flying the actions".
I'm sure there will be a next generation of jet drones that are cheaper, but the autonomy kit here is probably still cheaper than new jet drones.<p>I think most fully autonomous jet drone programs are also designed to work with a main crewed fighter jet that coordinates them, so there is also a benefit in getting the fighter jet to be able to fly better than a pilot.
More like it'll fly itself while the pilot engages in other tasks.<p>Every pound of weapons officer/navigator is a pound you can't put on target all else being equal.
* You're absolutely right. My altitude is dropping precipitously. Let me check my elevator position to see if that's causing the problem.<p>* Bash Check elevator position<p>* Thought for 8s
"I apologize, that was indeed a wedding party, not a terrorist camp I wrongly assumed. You are totally right, wasting those 3 Hydras was unacceptable. On a brighter side, we've got 39 more to go!<p>Should I seek another target looking similar to the one I unintentionally destroyed so as to tune my decision-making, or give you an advice on how to wipe the traces of this mishap from the mission data recorder?"
So like some sort of automatic pilot then?
Conventional autpilot is like a zip tie on the throttle, only it deals in outputs (your actual heading) rather than inputs (pedals and stick)<p>I assume what they're going for here is something substantially smarter and more reactive to changing conditions than that. Something that can be used during the "unscripted" parts of the flight during which a) you're most likely to have multiple crew working hard b) you're most likely to be pushing the limits of what one crew can do.
Does the presence of humans on board limit the maneuverability of these aircraft in any way? E.g. in terms of acceleration?
Drones have human operators. This is a murderbot.
The drones of Ace Combat 7 are coming closer to reality
Not a great thing for aggressive state prone to electing nutjobs and supporting genocides to have automated killing machines that will not even get PTSD afterwards.
Hey look, the lawn darts are now AI-controlled.
[dead]
What a waste of resources.