Navy F/A-18 squadron commander's take on AI repeatedly beating real pilot
thedrive.com
thedrive.com
A great book that rings more true than ever talks about the obsolescence of weapons systems. I believe it was Freedman[1] A key thesis is that as a predominant weapon system becomes more vulnerable, it becomes more expensive, as it’s burdened with more defensive measures to survive from cheaper more expendable systems. In World War 2 the canonical example was the battle ship and the aircraft. Now it’s the fighter jet and the missile (or drone) etc. Just look at the ridiculous state of the JSF.
Manned missile trucks are valuable for countering electronic attack, but not dogfighting. Still, practice dogfighting is one of the most fun things you can do, too bad it hasn’t really been applicable since Vietnam. This post from the other day reminded me how fun it is and seems like a great fit for VR.[2] Onward last Samurais! Just try not to spend all the taxpayers money on the way out.
[1] https://www.amazon.com/Future-War-History-Lawrence-Freedman-... [2] https://news.ycombinator.com/item?id=24255191
Cost is a huge problem. We now have unmanned carrier launched refueling trucks that I bet are incredibly cheap. Make them a tad bigger, put missiles in them, and you have a super cheap replacement for the F-35. The F-35 can stick around in small numbers for when you really need a human, if you really need a human in the cockpit. Cost has to drive us to this.
https://www.defensenews.com/naval/2018/08/30/us-navy-selects...
I don't understand why people persist in the fantasy that removing pilots will make aircraft cheap. That doesn't align with reality. There are some cost savings but when evaluated on a full program lifecycle basis they are relatively small.
They’re paying Boeing to learn how to build them.
F18s cost only a little more than a third of that per unit.
It's a bit like saying a soldier doesn't need a bayonet / combat knife. 99.99999% of the time yes, but would you bet YOUR life on it?
The military industrial complex is absolutely a beast that requires oversight and restraint. Just as the intelligence complex does. I wish more people understood and appreciated it. It doesn't need to be dismantled by a long shot necessarily, but BS needs to be called.
It probably depends on the situation and what you do. IIRC, missiles lose effectiveness at the kinds of short ranges where you'd use guns. In some situations, running might put you in a more vulnerable situation while closing to a gunfight might even things up.
If they try to extend out for a better turn rate you hit them with the missile. If they turn too hard they lose energy and you can gun them. The fights tend to end earlier because it's so much easier to make a mistake.
Also with missiles, countermeasures can be employed which are pretty effective. Can't fit much brain in a 40G missile that is mostly solid rocket and explosives. Missile warning systems and automatic countermeasures are remarkably effective. Not so much against 30mm
Yes you're probably going to die, but at least you are not a sitting duck.
They even give pilots a pistol or a sub-machine gun in case they eject over enemy territory. It's probably dead weight, but if it can save someone once, why not? I wouldn't go without it even if intellectually I know I'm not going to take down the whole enemy army with a 15 shots pistol.
Particularly Growling Sidewinder has a lot of great videos and I can't stop watching them.
A great analysis by Mover, a real pilot and DCS player.
I can play games where real-world motion matches in-game motion (Beat Saber, PokerStars VR, Superhot VR, Zombie Training Simulator, Space Pirate Trainer, etc.), but any sort of driving or flying makes me sick. Even games that make me use the controller to walk long distances make me dizzy and nauseous.
Then one day I was flying, and then I was flying and then I was still flying 4 hours later with no sickness. It hasn't come back since.
And the analysis of this incident: https://www.youtube.com/watch?v=yy16Bv14I-c
Having read some some history, the obsolescence of weapons systems is indeed a cultural problem that people become unprepared for in the euphoria of advancements. Classic example being Syracuse vs Rome, where Rome was held back by the weapons of Archimedes only to be sacked after the Syracusans decided to have a party and left a gate open.
A bit tangential but I'm curious. Can you enlighten me, or point me to where can I read more about it?
Also keep in mind that these are billion-dollar programs, so expect to get a lot of FUD and propaganda from the government contractors building them. I remember riding the DC Metro and rolling my eyes when I saw ads for the F-22 at Metro Center and Dupont Circle.
But the ability to control the plane is the minimum capability of an AI. And in combat, controlling a plane includes the capability to deploy countermeasures and take evasive action to avoid getting shot down, and defend itself to stop an attack.
I doubt the purpose of this was to create a great dog fighter... but rather it was to prove it's minimum capabilities work so they can continue developing it to operate in more dangerous/common situations... like taking out ground assets in the initial wave; or adding it to existing drones so they can take some defensive measures automatically; etc.
To me it seems that almost all wars of the past decades transformed into: “irrespective of your weapon superiority - it will all cumulate into asymmetric partisan wars that drag on for years and cannot be won”.
In the end it almost feels like “whoever drops more ordinances on the civil population to irrevocably break their morale may come out as a winner a bit more”.
How can wars be realistically be won today?
You'd be better off from a cost perspective running a swarm of hundreds of drone strapped with explosives, only one of them needs to get lucky to disable a multi-million dollar plane and the drones themselves can be pretty cheap.
Brannigan: "Killbots? A trifle. It was simply a matter of outsmarting them."
Fry: "Wow, I never would've thought of that."
Brannigan: "You see, killbots have a preset kill limit. Knowing their weakness, I sent wave after wave of my own men at them until they reached their limit and shut down. Kif, show them the medal I won."
Have you not just described an anti-air missile?
Also, drones are far slower than a missile, and even far slower than the airplane they're trying to target.
If you knew where the plane is headed, the question could become: can you create a wide enough curtain ahead of time that can home in on the incoming plane? Chasing is not needed that way.
So that's a big if.
Plus, the kind of small drone you’re describing can’t operate at the altitude of a fighter jet. As the air thins out you need a lot of speed to keep sufficient oxygen flowing into the jets, and non-jet aircraft can’t operate. So unless the drones are going to be blimps, a drone that can climb to fighter jet altitude is going to be pretty big and look a lot like a fighter jet.
Airwar is not about who shoots more planes, it's about who will blow up enemy's airbases first.
Animated example of a convoy getting hit by lasers:
Live example:
Energy dissipation of a laser in atmo is going to require crazy amounts of energy to go even a moderate distance, so either those lasers are being fired from something really close to the airfield, or else you've got some S.H.I.E.L.D style flying aircraft carrier thing.
A weapon without munitions, is a useless weapon, even if it costs $300M a pop. A war is won by a soldier, and lost by a logistics officer.
The maximum what a few fit men can load very slowly, unassisted is 250kg, and usually only 100kg by a single man. Gravity fed refueling is also quite time consuming.
Why do we have air-superiority fighters? It's to shoot down any aircraft that might be harassing our ground units, and to ensure that our own CAS/attack aircraft can freely harass their ground units without opposition. An F-22 with an air-superiority loadout is pretty much useless in the ground war. So just let it fly around, and if it shoots down 8 drones with its 8 missiles, big deal. There's hundreds more where those came from. Meanwhile any aircraft that are actively strafing, cluster-bombing, etc. the ground forces have to fly straight through the drone swarm, with likely pretty cataclysmic results.
Now which of these mission profiles can destroy an infantry company more cost-effectively? The JDAM may blow up a vehicle plus any infantrymen around it at the cost of $20K. The drone can individually target each infantryman with a bullet from a range of ~50-100 ft, for perhaps 100 kills for a $1000 drone, and it can be tactically controlled by other infantrymen on the ground, with instantaneous communication, to respond exactly where it's needed and avoid friendly fire.
If drones are going to be used for CAS, then they need a control channel or Skynet AI. Any control channel in use will depend on LOS, and will be jammed to oblivion in a real war.
I think there's an impedance mismatch between some folks who assume "countering" an F-22 or B-1B means shooting it down, and others who assume "countering" it means making it cost-ineffective to achieve your strategic aims. For the latter folk, having your manned aircraft constantly plink away at your forces and shoot down your drones is still mission-accomplished as long as your drones are doing a lot more damage than your forces are receiving.
The aircraft-carrier vs. battleship debate is again relevant. The two times that an aircraft carrier and a battleship got into a direct fight (HMS Glorious vs. Scharnorst, and USS Gambier Bay vs. IJN Yamato), the aircraft carrier got reamed, and sunk in short order. But this didn't matter, because aircraft carriers and battleships almost never got into direct fights. There was Pearl Harbor, Taranto, and Malaya before the belligerents realized that sinking battleships was pointless, and then there was Sibuyan Sea and Okinawa when there were no more targets left. In between were a long string of carrier vs. carrier battles where the battleships were used as AA escorts or moved to shore bombardment & convoy escort duties. Similarly, why bother countering a few smart bombs when your drone swarm terminates every enemy soldier on the battlefield within an hour and makes any attempt at interference suicidal for manned aircraft?
Since we're being realistic, you could use the launcher I sketched out as an elementary school student. It had missile tubes on every surface, including around the wheels.
No one's using tactical nukes against a drone swarm.
If a drone swarm proves lethal to a multi million dollar aircraft of ship, tactical nuke could be a suitable response.
Unlikely, sure, I'm just saying though....
The Genie anti-air nuclear rocket [1] was designed to take out soviet bombers, in a time when guided missiles were too immature to provide the required precision. A more modern guided nuclear anti-air missile (with features such as PAL security and tritium boosted fission for a higher / dialable yield) would probably work quite nicely for reducing swarm density.
The way I see it, we're probably headed towards a cold (at minimum) war between sino-russia et. al. and the western allies, so (in the absence of something more effective) tactical nuclear weapons to counter aircraft swarms will become more tolerable as tensions continue to rise - far better to use one to take out a drone swarm and cause drama than risk losing an air wing (or something else that interferes with force projection). Hopefully I'm wrong though.
If you can develop a drone swarm that can totally own an airspace then chances are you can develop drones that can beat them.
There's no reason either of these needs to be true. A missile with an active control channel is a drone, and another class of military drones is regular manned aircraft converted for remote and/or autonomous operation, like the QF-16. A purpose built air-to-air combat drone could have a wide range of potential performance envelopes, but would almost certainly both have speed and maneuvering advantages over any manned aircraft it was expected to target (quite possibly a lot less range, though.)
They do if you buy enough of them.
Yes, but a loitering anti-air missile.
Just for fun, and I have no qualifications whatsoever, but the Aim 9 comes in at 188 pounds (85.3 kg), add say 20kg for a release mount? (just guessing), so might be doable.
Might need something more upmarket than a Griff 300 (max altitude 1000m - see https://www.dronetechplanet.com/the-griff-300-review-drone-t...) but even with that, if you had network of these remotely operated they could pack a hell of a surprise for an unsuspecting pilot (missiles would only go active when released), can be positioned strategically around assets.
You can have a wall of evenly spaced drones with explosives. When an aircraft passes all adjacent drones explode, the shockwave damaging the aircraft.
A wall of drones 100 feet apart, covering 10,000 x 10,000 feet, is 100 x 100 drones, or 10,000 drones, which is a wall that's a little under 2 miles by 2 miles. That's a lot of drones to cover very little area, and they have to each carry enough explosive to create a shockwave that might damage the plane.
Moreover, this is a one-shot weapon. Once something sets it off (like a plane-like drone), there's a big gap in the wall. The idea of saturating an area with explosive drones has a scaling problem that would probably make it cost-prohibitive quite quickly.
Of course drones lack the biological limits on G-Force and the economic scales and factors like pilot deaths mean the situations are not identical.
The real question to be answered is probably one of "expendible drones vs premium super drones" long term as the age old battle of quality vs quantity.
If you're going to use the drone in an A2A role then it needs a powerful radar for accurate targeting. Targeting from other platforms using data links is possible in theory, but in practice only works in a narrow set of ideal conditions. Modern radars and other sensors are themselves super expensive.
I'd expect to see a shift toward more visual and IR sensors as well, to combat stealth technologies. Modern ultra-HD cameras + image recognition algorithms are a lot more sensitive than human eyes are, and you can field many more drones than AWACS aircraft with much less cost of a loss, letting you bring the relevant detection platforms in closer.
Sometimes I swear there is a pas de deux going on between these apparent enemies.
Once a war is going on the you can move to a much more aggressive strategy. The US can just declare that any boats within a certain region are subject to summary attack and all shipping should steer clear. Attack aircraft can start sinking these small boats as soon as they leave harbour (or most likely cruise missiles blow them up at anchor).
"There has been only one dogfight involving a U.S. aircraft in the last 20 years: in 2017, a U.S. Navy pilot shot down a Syrian fighter." [1]
So you have to wonder a bit how much time is worth investing in that aspect of these planes.
[1] https://www.wsj.com/articles/duels-among-the-clouds-11565878...
https://en.wikipedia.org/wiki/Close_air_support
Without close air support, troops on the ground can't overwhelm enemies with firepower. Without air superiority, we don't provide close air support. Our strategy dictates air superiority to enable our tactics of close air support.
The reason there's only been one dogfight in the past 20 years is not because these fighters aren't effective, or because they're a relic of the past. It's that they're too effective. We own the sky and nobody is capable of challenging us.
That calculus is different where combatants are more closely matched (e.g. India and Pakistan), or it may change as opponents improve their capabilities.
Also per unit costs are misleading. You can’t easily quadruple Or 10x your number of air bases, number of trained pilots and air crew, number of aircraft carriers to carry all your cheap and cheerful swarm fighters.
A lot of fast jet economics is driven by getting maximum utility out of deployment capacity and support infrastructure that is itself irreducibly expensive.
As for other countries: it’s still in your geopolitical interests to have countries who aren’t allies using your gear - not just for the money - but because it means you already know their weaknesses and you can sell them nerfed “export editions” to ensure you have technical superiority and to lock them in to buying parts, accessories, training, and upgrades (and successors!) from your own nations’ defence contractors and you can saturate a market that your geopolitical rivals (Russia, PRC, Iran, etc) would otherwise have access to, thus weakening their sphere-of-influence.
For a recent example of it going wrong, see the S-400 sale to Turkey.
Arms sales are an ugly thing - but if NATO states didn’t sell arms to Saudi Arabia, Libya, Turkey, etc what would the alternative be?
Education and health care?
I wish - I really do :/
When countries set their budgets for the year - they'll set aside, say, 15% for defense - if their DoD reports to the cabinet that none of the superpowers will sell them new toys then that government's cabinet is not going to suddenly craft a new budget and divert that money that would have gone to new fighter-planes or small-arms to other civilian areas - that money will certainly stay with their military who will find some creative use for it.
...and that's the best-case scenario - for a country that isn't expecting hostilities.
What I meant by that was that if NATO doesn't sell to a third-world (using the strict/classical definition of "third-world") nation who is looking to buy arms then a rival power or superpower will - they may even give-away some amount of arms gratis to sweeten the deal and to "thank" the country for leaving or even reconsidering the NATO/OEDC sphere-of-influence - obviously Russia isn't the Soviet Union and they don't have the same expansionist goals, but I believe we're seeing the PRC making its moves. The PRC hasn't yet fired-up its export economy for defence gear yet - but I believe it's inevitable.
I stress that I don't want to start-up any Sinophobic sentiment - but what I'm seeing the PRC do is exactly the same thing I'd expect to see from any rising superpower looking to preserve their unending explosive economic growth... history doesn't exactly repeat itelf, but it often rhymes: the PRC wants influence, apparently to benefit itself (and I am concerned about the signs of Han ethnonationalism being the motivation behind their foreign and domestic policy objectives - and the PRC doesn't stand for the same principles I do) - I'm concerned where it will end-up. So if it means having to sell arms to countries (as part of an socio-economic-military package) to keep them in NATO+OEDC (which _generally_ value the things that I value) with the view of keeping them out of the PRC's SoI then that's something I'll have to accept.
When I think of the SU’s expansionist goals I think of Seven Days to the Rhine rather than territorial gains of convenience - so I apologise for being vague in my original comment.
I know the same argument applies to the PRC: their occupation of the South China Sea is opportunistic and strategic and so is more comparable to Crimea - it’s not like China is invading Vietnam or South Korea with intent to annex them.
The USSR is dead and gone and no one in power in Russia wants it back, save for pensioners who had a better deal back then.
Putin wants the Russian Empire back, which held Poland, the Ukraine, most of the Baltics, etc.
Why be a Premier when you can be a Tzar?
It is weird that we sell F-15s to Saudi Arabia.
[1] Iraq 1 & 2
[2] Yemen
Also the jets sold overseas are generally "export" quality and have downgraded features or require embedded expertise to maintain.
My takeaway from that data point is that the Syrian military might be very interested in such an investment.
https://en.wikipedia.org/wiki/McDonnell_Douglas_F/A-18_Horne...
The unclassified ceiling is 50,000 feet, with a 250 m/s climb rate. Could a drone strapped with explosives come anywhere near that level of performance?
Why do you think fighters carry a bunch of “drones strapped with explosives” (air-to-air missiles) for air-to-air combat?
The thing is, when you optimize those, they tend to be comparatively short range and have no loiter time, so it's nice to have something to carry them into an engagement area and, if you need to patrol or escort, it's nice for it to also be something with some loiter time.
They don't typically carry _hundreds_, though. Carrying capacity for air-to-air missiles is pretty low. You get only a few shots.
So if an enemy ship, perhaps an aircraft carrier, really pissed off a single US destroyer, it can decide to send all hundred supersonic anti-ship missiles in from the same direction at the same instant of arrival, or from every point on the compass, or maybe three directions at once or who knows what kind of crazy four dimensional incoming attack pattern, like waves of ten missiles at a time every second for ten seconds from ten different directions. So how do you distribute your drones so as to soak up a hundred incoming from one direction while not getting sneak attacked by say five from every 30 degrees on the compass at the same time?
Its like say you have a Greek Phalanx, that's pretty well defended over a couple hundred feet, but try to use the same one phalanx spread out to defend all one thousand acres of city-state farmland simultaneously and its not going to work out well.
So just like the famous rocket equation, you'll find that the design space narrows down to drones that are no longer cheap or casually expendable - essentially supersonic capable cruise missiles with loiter times that tend to cost millions of dollars a piece. Certainly cheaper than manned aircraft but not something that can be used in the hundreds casually per engagement.
The other option is of course what's done today - strapping a relatively cheaper air to air missile to an even more expensive but fast and survivable aircraft which brings it to the war zone and gives it kinetic energy at launch.
Then when not engaged in air defense, these drones engage in CAS. Taking down fighters is a secondary mission - their real role is to dominate the battlefield, and part of that is by making it so unsafe to enter the airspace above the battlefield that it's futile suicide to try.
This is without factoring that a typical war-fighter could simply shoot through such a passive defence wall from dozens of km away if they really need to and if they are moving at Mach 2, they'll be through the hole before it reforms and because of lack of speed, there's essentially no chase capability once through.
These kinds of ideas while interesting remind me of the Magniot line and we all know how it fared
The rest of the F-22's range and service ceiling, you don't care about. You let them fly around all they want. If they want to go fly around at 50,000 feet at Mach 2, have fun. An F-22 is basically harmless to a drone: it can shoot down 12 drones with its 12 AA missiles, each of which costs > $1M, about 1000x what a drone costs. There's no pilot in a drone whose life is sacred, and there're plenty more drones where they came from.
You only need to engage the F-22 if they're coming down to engage ground forces. But then they need to fly through the swarm, at a huge risk to the airplane and pilot, and the swarm can react to put as many pieces of plastic in the F-22's path as possible.
If the F-22 stays up at 50K feet and drops precision-guided weapons from 30 miles away, they lose the attrition war to the drone swarm. A drone can very accurately put a bullet into each infantryman and a shaped charge on top of every tank with little resistance, unless the enemy is itself fielding autonomous drones. An F-22 can put 8 SDBs on 8 targets. Which is more efficient?
Sure simple small drones can attack ground targets but that's a different problem from area denial to air superiority fighters and it assumes you already have a ground presence (ie. insurgency ISIS style) to get within range where you minimize risk of detection and flight time. If detected, an ISIS style drone can be shot down with a rifle.
Again I'm not saying pilotless attack / defence is not going to come - it obviously will. I'm saying it probably won't be the tens of thousands of cheap drones approach for the reasons above - it will be faster, more expensive drones which will be a significant fraction of manned aircraft cost.
That was perfectly visible in the recent Syrian Civil War. Once the Russians decided to actively step in the conflict sure that they sent a rocket or two directly from their bases in Southern Russia/Caucasus, but I suspect that was more for propaganda reasons.
Soon enough they set up an air-base for good in Northern Syria (only a few tens of kms away from rebel lines) and they also brought in a battleship (or whatever is called nowadays the ship from which you can send rockets) off the coast of Syria. I'm pretty sure that proved a lot more cost-effective for them.
Later edit: A news article from that same timestamp [2] of Iraq complaining that it hadn’t been warned by the Russians of the latter sending missiles flying over their heads. It also includes a video of the supposed launch.
Later later edit: Here's a (short) Twitter video of one of those missiles flying over Northern Iraq [3]
[1] https://www.bbc.com/news/world-middle-east-34465425
[2] https://www.washingtonpost.com/world/iraq-says-it-was-not-wa...
[3] https://twitter.com/NewsOfIraq_EN/status/668192312497950720
"Guided missile cruiser/destroyer/frigate" would be more correct terms. Frigates are rarely employed in land-attack missions, and only 2 or 4 countries have cruisers (US + Russia, as well as China and South Korea if you count their respective oversized destroyer classes as cruisers ). There are no battleships in active service in any navy today. The Russians do, however, possess the closest thing: the world's only nuclear-powered guided missile battlecruiser: https://en.wikipedia.org/wiki/Kirov-class_battlecruiser
I dont think thats a problem that can be solved today to the exclusion of manned aircrafts. If you are a terrorist it might not matter but for the kind of nation states that can field fighter jets it certainly matter.
That discussion about swarms always devolves to "just push human beings near the action before launching the pitbull outta control killer drones" which for the last 30 years has been repeatedly successfully been implemented as modern jet fighters carrying short range sidewinder IR guided missiles, which annoys the people who thought they invented the new idea of explosive drones, little did they know we've been doing that for generations and its unfortunately probably even more expensive than swarms of WWII era P-51s with human pilots.
While I do not believe I am at liberty to provide details (even lo so many years later), I was witness to the first use of (arguably) VR to prototype and introduce a new weapons platform into a combined arms battlefield simulation. The short version is that on Monday morning, despite all the deep thinking done by smart people about how this would all work out, none of us came close to predicting what turned out to be the net effects of the new system as seen by Friday. I was there in my capacity as a nerd simply to keep the blinking lights blinking (vs the capacity of a combat domain expert), but watching the whole thing unfold was a mind-blowing demonstration of the "Law of Unanticipated Consequences."
None of us are as smart as we think we are. We are no smarter than our adversaries. The world is more complex than either of us can know.
Just fly in at a higher altitude and if there's some there go around them.
Drones carrying air to air missiles might be more challenging.
Regardless of the actual merits of whether dogfights are a thing any more, I feel like "because Top Gun" is not exactly a sound argument.
I did not pick up there is an argument about whether or not dog fighting still was needed.
The only "argument" I recall was the recap of how pilot deaths increased after training stopped, and decreased after training started. But it was a one off history lesson as part of the Top Gun school intro.
But the rest or the movie seemed very clearly - yes, the US needs dog fighting to defend its interests across the globe. There seemed to be absolutely no challenging of that idea within Top Gun.
Maybe I'm completely wrong (I'm not a pilot and have no military knowledge), but I feel like things have moved on since then in terms of both the technology and the types of wars fought.
Maybe someone more knowledgable than me can answer... when was the last time a US combat aircraft engaged in a real dogfight?
This isn’t to say there wasn’t propaganda, just that the propaganda was pointed in a different direction: for example a young idealistic police officer working with an veteran of the force to tenaciously investigate an elusive criminal gang for the glory of the Soviet Union. The focus was to make Soviet patriotism an attractive quality, regardless of being young or old. It was a lot more focused internally. In the meantime, a movie like Die Hard has none of that: German thieves that pretend to be terrorists are the bad guys, an all American hero is a good guy. American movies rarely have a true patriotic streak, as in “and this, son, is why we must work to better our union.” In fact, the idea of glorifying labor was so prevalent in the Soviet movies and so absent in American ones that this might be the biggest contrast. It shows a different face of patriotism and could easily explain the divide: some think patriotism is exercising your rights while others see it as holding back on exercising your rights. “It’s my right not to wear a mask” is a clear example of that.
Interesting, the closest Western execution of this Soviet theme, albeit much more cynically, might be "Child 44", which was banned in much of the FSU and bombed: https://en.wikipedia.org/wiki/Child_44_(film)
Personally I enjoyed it.
>>>American movies rarely have a true patriotic streak, as in “and this, son, is why we must work to better our union.”
"The Siege" might be one of the best for sticking to American values and trying to curb xenophobia. A big part of the plot is the importance of holding other Americans accountable for violations of our laws and our values, even when fighting against "foreign" enemies.
"You don't become a good communist by going to meetings. Or memorizing the manifesto. You do it with your fist."
Comrade Detective: "You're telling me the purpose of the [Monopoly] game is to drive your fellow citizens into poverty, so you can get rich? It's diabolical."
American subversive: "It's just a game. Nowadays the West finds it as a useful tool to indoctrinate young children into the capitalist system."
Not 80s, but Wolf Warrior is China vs. evil American mercenaries.
Many dozens of times over multiple decades in multiple theaters against multiple opponents, the vast majority of stereotypical NFZ engagements have ended in "two american planes primary and wingman have a brief dogfight against enemy aircraft to obtain ideal firing positions, involving visual identification, then end the fight with short range IR guided sidewinder missiles". There are, of course, occasional exceptions but the VAST majority of NFZ aerial combat stories look like this.
Its interesting that for 30 years NFZ ROE documents have been classified and are generally still classified and only rarely can you find an ROE illegally posted on wikileaks. HOWEVER the secret squirrel types seem to have ignored the internet and OSINT in general. So you can't legally find a piece of paper titled RoE describing how visual ID was necessary and BVR generally prohibited for all practical purposes, however, its trivial to read on wikipedia the stories of about a hundred NFZ engagements and then reverse engineer the classified ROE to match the large number of mostly identical individual engagements.
Or to summarize the above, the RoE for NFZ operations for the last 30 years remain classified, although the results of seemingly all air to air engagements have been publicly released, and traditional OSINT techniques applied to hundreds of "isolated" public anecdotes can easily reverse engineer the contents of those classified RoE documents, and the reverse engineered RoE documents pretty much mirror the fight scene at the end of the Top Gun movie; get close, dogfight into missile launch position while performing visual ID, end fight by launching short range IR missiles.
https://csbaonline.org/uploads/documents/Air-to-Air-Report-....
It's not a cloaking device, it's a <1 multiplier on the effectiveness of enemy radar. Both sides will still have radar, and detection range for a good stealth fighter with modern radar against itself is going to be higher than visual range. What stealth does do is it reduces detection range to well below the employment range of good missiles, which means that the first one to get a radar lock will either end the fight then and there, or at least be able to manouver into a very favorable position while the enemy has to evade a missile. This is why fights with stealth aircraft in simulation and training are so very technical and one-sided.
[1] https://en.wikipedia.org/wiki/Iran%E2%80%93U.S._RQ-170_incid...
In this case the AI is onboard, right? So much lower surface area for being hacked.
I hope you're right, but what makes you so sure? In a wartime scenario, if a country is short on skilled personnel why wouldn't they release AI combatants if they had the tech? Especially if they're facing a perceived aggressor or defending their territory.
Ha. I know of a device with a very simple AI program that can tell if an enemy is standing on it. If so, the enemy is killed. It’s called a mine. They have been used in every land (and naval) war since they were invented.
It’s a good point.
The level of complexity is arbitrary. Maybe the AI won't blow something up unless it has a heat signature. Maybe it has to detect a pulse. Maybe it won't blow up something that emits a certain beacon. Maybe the kill has to be within a certain timeframe. Maybe the target had to have a certain mass. Which of these considerations qualifies as AI?
From a different perspective you might think of guns and bombs as having similar problems—a human might decide to pull the trigger, but the human involved has limited control over who is killed and what is destroyed.
I think this has to be a pretty classic example of the way discourse can get derailed on HN, Stack Overflow, or the like—Alice has an interesting point to make, but Bob takes the floor to point out that Alice used a word incorrectly, even though he knows what Alice meant.
The discussion about landmines is actually a good point—the legal language around landmines has evolved somewhat over the years and you might hear them referred to as “LAWs” these days, which stands for “Lethal Autonomous Weapons”. So many of the the same treaties will regulate both autonomous drones and landmines, with the same language.
Words and meanings are not equally important. Words are only there to communicate meaning. You can communicate meaning without words, and you can even communicate the correct meaning with incorrect words.
There's an interesting discussion to be had here, but "are landmines AI" isn't it.
And yet, landmines kill an estimated 15-20,000 civilians each year. About one in five are children. All because someone decided to defend an area. You say this is “working as intended”, but—how can this be the intended result?
Automatic weapon systems allow you to make decisions about who lives and dies, and they let you escape responsibility by making those deaths occur at a different time/place from when and where you caused those deaths. If you were defending an area with teammates and friends’ lives on the line, you might be a little less careful about how your decisions affect the lives of civilians—especially the civilian deaths are uncertain and distant.
I’d go back and check with product on the requirements for that one. Sounds like the opposite of what I’d want.
I think maybe those landmines are supposed to kill specific targets, and avoid collateral damage.
So really what the OP means is the ethics of what to kill and what to let live should not be left to an AI.
https://www.thedrive.com/the-war-zone/25235/the-u-s-is-getti...
You don’t have a case of no human in the loop until that mine is capable of exploding or not exploding and it’s responsible on a case by case basis for deciding which to do.
United States, China, Russia, Saudi Arabia, Singapore, India, Pakistan, United Arab Emirates, Egypt, Iran, Israel, Vietnam,North Korea, South Korea, Armenia, Azerbaijan, Bahrain, Cuba, F.S. Micronesia, Georgia, Kazakhstan, Kyrgyzstan, Laos, Lebanon, Libya, Mongolia, Morocco, Myanmar,Nepal, Syria, Tonga, Uzbekistan.
Those countries together represent a significant portion of the global population!
https://en.m.wikipedia.org/wiki/List_of_parties_to_the_Ottaw...
Or further still, doesn't the camouflaged trap, or non-attendance-required trap of any sort in general, become the original autonomous tool?
This is unfortunately completely wrong. Fully autonomous weapons have already deployed for years, and are currently in use on the North/South Korean border. Fully autonomous weapons systems, including drones that use AI to autonomously choose and destroy their targets have been under development for many years.
https://www.nato.int/docu/review/articles/2017/07/28/autonom...
There's a reason experts are sounding the alarm - because the ethics of killing are being shifted to AI right now.
>Kallenborn, an expert in unmanned systems and WMD, describes one type of swarm that he calls an Armed, Fully-Autonomous Drone Swarm, or AFADS. Once unleashed an AFADS will locate, identify, and attack targets without human intervention
https://mwi.usma.edu/swarms-mass-destruction-case-declaring-...
As for taking control of an AI drone, you can eliminate that possibility by only allowing abort orders remotely.
Classical drones still require a remote human. This is a different league.
You're correct that flight computers are generally required to keep dynamically unstable aircraft flying -- but that's done at a much lower level than anything that's connected to a radio.
Other have already answered this, but I want to chip in with my own experience working for a submarine and military boat manufacturer + talking to navy officer.
They already have system to acquire and try to distinguish friend of foe. The only reason why the system is not pushing the launch button by itself is because there is a lot of tactic involved in starting an engagement. Especially for submarines who might prefer to remain hidden than launch a torpedo which will uncover its position. Also, those system are not perfect, one navy officer who was on radar duty told me that he activated the alarm on his ship after seeing something on the radar that ended up being a flock of seagull.
If the system was able to manage tactic and be able to distinguish friend of foe reliably, they would give it the power to kill.
As the author mentioned, the dogfighting AI seems to counter the real pilot’s moves effectively while maximizing the amount of energy maintained. It seems to have caused the real pilot to overshoot or undershoot while maintaining or gaining energy. The countering also included adjusting aim to hit the real pilot as the real pilot tried to maneuver away. It wasn’t just perfect information or efficient flying.
https://www.popularmechanics.com/military/aviation/a29847417...
https://www.uasvision.com/2014/06/05/target-qf16-could-becom...
1. The AI had "total situational awareness" (perfect position, direction, velocity of opponent). Even given 360 cameras, radar, etc there is still uncertainty, errors, etc.
2. The human was an expert pilot, not a sim expert (see videos, he was going way too fast several times, not something he'd do in a real plane). A DCS master would likely have done better.
3. The AI wasn't trying to avoid head-ons.
With this and the Starcraft2 blink micro "win" I'm just going to these from now on.
Why should it? Presumably the AI pilot is backed up / remote / not precious - while the human pilot has an inherent value - isn't reckless disregard one benefit of automated systems?
This was testing the AI. If real-world AI was going to be deployed in warfare shooting down manned planes, you'd make custom hardware that is smaller, faster and 10-100x cheaper than a f-35.
For some reason, we can’t justify more on civilian research but the military easily can, if it embraces it as part of its future.
- If you don't, then the enemy will
- If you do, then you're inviting the enemy to do the same
So, without trust and mutual agreement on an international level, this is a self-propelling premise that cannot be stopped.
Game theory 101.
Some might ask the same question in the context of human children.
They may give troops on the front line some leeway, but within parameters for pursuit of a goal.
I don't think Defence wants "uncontrolled" anything. Even if they manage to develop fully cognizant AI they will want to know the exact parameters it can work within.
Of course, humans make errors...
Aside from, like, the guns 'n' stuff.
Nothing to worry about, move along.
I think recent events have clearly demonstrated that biological weapons are much much cheaper.
create a killer virus
inoculate your people
release it at the rest of the world
profit?
So if it gains autonomy there is high chance of it destroying everything else before dying itself.
I found the story of Horizon Zero dawn highly plausible. The main takeaway - dumb military AIs destroy all the living things. The only bit we are missing is fully automated fabrication.
Supply chains for high tech goods such as AI-powered autonomous fighter jets are extremely complex. We've already automated some of the individual steps, and will surely in the coming decades automate many more. But high-level orchestration of the chains as a whole, that requires many human-level intelligences to cooperate in advanced social systems, I don't think anything short of general AI could replace that.
Not only what we are missing is self replication but everything else including the important part and what was the problem “converting bio-mass into energy” without that it wouldn’t be an issue, not to mention that HZD makes little sense in general even with the coverup nukes exist and they blow up shit just fine their lore made little sense in the long run as they couldn’t replicate they just had really big robots the octopus ones that could manufacture more of the smaller ones.
And more importantly since the biomass conversion was basically nano machines why couldn’t you produce a swarm of them that simply eats robots? Or why couldn’t you produce a swarm of robots to fight the rogue swarm.
The whole “unhackable” thing also doesn’t makes sense and if they were then well make another swarm that just is unhackable and let them duke it out. The technology and more importantly the consistency of the technology in that universe makes absolutely no sense.
As much as I liked the game the story while cool was about as plausible as an alien invasion that triggers a zombie apocalypse followed by the second coming.
Except you know, it kills people.
Yea, and the game is life and death. A military minded AI won't distinguish between surrendering and fighting enemies, won't distinguish between enemies and non-combatents, might confuse a shutdown order with an electronic attack. Just cause AI isn't by definition "smart" doesn't mean its harmless.
"mostly" still leaves a lot of room for bad actors, but these sorts of compromises have worked surprisingly well enough that the failures were notable. Political problems are only going to have political solutions, and almost by definition those are imperfect.
Frankly, we shouldn't be developing AI outside of institutions with STRONG ties to humanist missions (possibly including religion). AI is not for war, and it's not for "efficient" resource extraction and development; it's for helping humanity to safeguard itself and its potential.
It has completely failed to keep nuclear weapons in check, imo.
Keeping AI technology out of existence has the same problem as keeping nuclear weapons out of existence - the incentives are too great for any nation to have them.
If a nation is armed with nuclear weapons, the only possible check on them is another nation that is armed with nuclear weapons. Thus even in the existing non-proliferation framework, the largest nations (the United States, Russia, China, several European powers) ALL still have nuclear weapons. The reason why is that if any nation was the only nuclear armed nation it would be impossible to defeat with any quantity of conventional force.
The same argument applies for AI. It would be impossible avoid the existence of militarized AI if it provides the same level of force multiplication as nuclear weapons. It would be necessary for any nation to try to gain the same technology as a deterrent against attack by a rival that possesses the technology.
Really debatable. Bribe enemy generals, drop commando paratroopers right on top of enemy's political leadership, even at 90% attrition rates, bomb vital military infrastructure, like strategic stockpiles, to demoralise the enemy (making them understand that the war will be long, hungry, and ugly,) even if it means sacrificing the entirety of your airforce, and wait until your land force finally reaches enemy territory amid this complete chaos.
It seems like if what you are arguing is possible, nukes don’t need to be invoked at all. Why wouldn’t you use the same technique against a conventional enemy, if it is possible at all?
There are exactly 0 countries on record who went on offensive against nuclear states, with intention to invade them.
To execute a strategy like this, one needs to be quite a monster to rationalise such losses in their sane mind.
It is as ruthless to your own soldiers, as it is to the enemy: forcing a stronger enemy into the most ugly, meat grinding form of ground war, with exact intention to exact maximum attrition losses, while negating advantages of high tech weaponry. Sending them a message "All your bombers, and missiles are useless at the bayonet range, you will have to bury us with your bodies"
Even Mao only sent covert cues to Soviets that he is ready to repeat Zhenbaotai, but even he had no real intention to follow up on them. That was a pure theatre.
It is insane to go to war against a nuclear armed opponent. Sure, you could kill half your able bodied population of military age and maybe win (what if something goes wrong in your battle plan? Now you're nuked and the opponent is still combat ready), but now any of your neighbors can easily take advantage of your now devastated country.
You can't bomb much of vital military infrastructure if your own infrastructure - not only airfields, but also population centres - have been nuked. Your bombers can drop a single full payload, after which they will have no bases left where they can return to reload. I mean, we have Cold War studies of all of that - the only way to ensure that bombers really have the ability to bomb enemy infrastructure was to ensure that they are up in the air patrolling, or alternatively scramble to launch within minutes of an alert, since an hour after the start of WW3 you only have a single strike with the airforce that you have in the air.
You can't demoralise the enemy by "making them understand that the war will be long, hungry, and ugly" because the war will end in a single day, after which half of your population will be dead, and the other half in need of food, electricity and supplies. Sure, you will have people willing and able to defend the country, but nobody will be invading; but after being on the receiving end of a nuclear strike your ability to deploy and supply a massed overseas invasion is destroyed, so you're not a meaningful threat anymore, the conflict is over.
Nuclear deterrent is the ultimate defensive weapon because it ensures that any war is unwinnable for the attacker - if the attack fails, then it fails, but if the attack succeeds, there's no good response to an ultimatum to accept ceasefire or have your cities nuked. All your soldiers and commanders know that if they invade and win, then their families and their homeland will die anyway.
And I read much of them, and what I wrote above is a restatement of what I read.
> Your bombers can drop a single full payload, after which they will have no bases left where they can return to reload.
Yes, but even this will likely be enough to bomb biggest C3 centres, bases, stockpiles of food, ammo, and fuel. Just enough to ensure 1.) initial shock, and C3 paralysis, 2.) making sure that the adversary will be running on the leased time too.
Even at the height of cold war, each side had at most 10 strategic level stockpiles. Without them, the huge force would've ran out of supplies from regional bases in under 6 month.
> after being on the receiving end of a nuclear strike your ability to deploy and supply a massed overseas invasion is destroyed,
If you have your entire tank, and mechanised force on the go, and one leg on foreign soil already. It doesn't matter. The enemy will either have to nuke own cities captured by your force, or contest them with conventional force. Similarly, they will have to either set their equally diminished supplier on fire to deny them to invader, or have to contest them.
> All your soldiers and commanders know that if they invade and win, then their families and their homeland will die anyway.
Yes, it requires one to be a madman to accept such "victory." Especially if you already are a somehow major power, as you know that in today's free-for-all world, a third power will swoop in to claim spoils in no time. But think of it as a "counter-MAD," or "asymmetric MAD," or "madman's MAD"
Genetic engineering on humans.
Politically there is a ban. But given your "not perfect enforcement" and 2000 billionaires I guess at least one of them has this goal - whether it is considered kooky or not. Someone is trying to engineer their offspring for intelligence/looks/health.
Whether they achieve it remains to be seen and will no doubt come out in history.
The first is done largely by the great powers agreeing that it's a bad idea from a military standpoint. That is, biological weapons are not stockpiled not because of all the obvious horrible consequences, but because the common wisdom is that they are ineffective weapons.
Keeping nuclear programs in check is mostly done through expense. Everyone who has them has made enough of them to feel safe, and doesn't want to make more because they are really expensive. After states have decided they don't want any more nukes, they sign agreements with each other to make sure no-one else is building a massive stockpile. If any of them wanted a larger stockpile, they would make it.
In general, powerful states can not be limited by agreements. They only ever work as a framework that allows everyone to do what they would have wanted to do anyway, while feeling confident of what everyone else wants to do and how to plan for it. The American obsession of thinking of the UN as world government and international agreements as laws really distorts your view on the world. That is simply not how any of this works.
But regardless of all that, even if broad international agreements would actually work to limit weaponry, it still wouldn't work, because AI is deniable. The initial AI isn't going to replace the pilot, they are going to augment them. And those are developed further until the pilots are no longer needed, and past that. So you can develop all the way to the point where humans have become a detriment, without needing to show your hand. And only deploy completely unmanned platforms once the proverbial shit hits the fan, which is after the point where everyone has stopped caring about international agreements anyway.
Can you point to any evidence that this is the case? It doesn't sound plausible to me.
This is about chemical weapons but a lot of it applies to biological as well. Biological weapons also have the added benefit of possibly becoming a global pandemic and infecting your own military and economic base.
I don't know if it's because of Trump, but as far as I can tell Americans don't think of the UN at all anymore.
At all. There have been so many recent events where the UN could have been involved, but wasn't.
E.g. when the USA recently smoked General Soleimani with Hellfire missles, where was the UN?
I watch local and network evening news on a regular basis, but didn't hear anything about this being discussed in the UN.
Perhaps it was debated, but it went unnoticed by about 99% of the US population.
This is a good example of the weird attitude where Americans think of the UN as a world government.
UN is a place where diplomats of different countries keep constantly in touch and maintain relations. That's all it is. UN does not have opinions. The member countries have, UN is only a place where they are presented. UN does not do things. The member countries do things, UN is the place where this is organized.
It amuses me to think that Aperture Science had a protocol for rogue AIs with the codename "Birthday Cake". GLaDOS did not understand why all of the humans kept talking about cake as they died. They really seem to like cake...
Then in 1969-73, Senate majority leader Mansfield decided that this research might be being misused to fund science in ways that wasn't directly useful to fighting Soviets, so ARPA became DARPA (the D is for Defense!) and now researchers needed to explain why their science might help build missiles.
Which is prima facie untrue. While there are benefits to basic research over the long term, it took roughly 60 years from the discovery of the laser to the beginnings of deployable laser weaponry today. Those sorts of timeframes aren't that useful for a military expecting a conflict in the next few years.
Only if you have little knowledge of military.
> it took roughly 60 years from the discovery of the laser to the beginnings of deployable laser weaponry today.
But it took less than a year for it to be used by the CIA and FBI for spying (an interferometer was one of the first applications) and not much longer for it to be used for range finding and targeting for the military and not much longer than that to be used to determine mass of objects.
The directed energy weapon you are referring to isn't the only application that the military, CIA, FBI, and other agencies have for lasers. There are hundreds more than what I mentioned and a dozen of which were in practical use within the first 5 years of its discovery.
Just because you aren't aware of the uses of such technology doesn't mean that they aren't used and doesn't mean they aren't __HIGHLY__ valuable assets. We could go on and include things like nightvision, underwater communications (other types of communications as well), guidance, or defensive applications but I think it is pretty clear that the laser is one of the _best_ examples of technology being used (and quickly adopted) by military/government, not worst.
That said, DARPA has always been one of the biggest funders of research in the US, whether it's basic research or applied. Just think of the DARPA Grand Challenge (self-driving cars in an off-road setting). While, yes, ultimately their goal is military application, this cultivation of research into self-driving cars is something that will benefit everybody, not just the military.
The reason to invest in civilian research while in times of peace is that you want products to go into the public domain and be mass produced. This means that the new fighters you can make are much cheaper to produce and can have much more advanced weapon capabilities. The only real difference between civilian and military research is if the knowledge gets into the public domain or not (with the exception for weapons, which is more niche than you'd expect). This is why DARPA does that funding, but it isn't just DARPA. The entire SBIR/STTR program exists to get companies to bootstrap themselves and mass produce things the government needs (yes, it is abused quite a bit). Many contracts exist for this reason too.
The last aspect is economic and cultural. If your country is doing well economically it is often able to avoid war in the first place. Rich countries often don't fight with soldiers and jets, they fight with computers and resources. Why invade a country when you can put economic shackles on it? China and the US are both doing this. The other part is cultural exportation. If your country is producing all the widgets that another country likes and enjoys then their people are not eager to fight you (as long as you're not being aggressive towards them).
All these, and more matter. War is much more than bullets and fighter jets.
This is __NOT__ what I claimed.
It goes both ways, and military applications create market pressures and timescales that wouldn't normally exist. Take for example, sustained fission, computing, or the semiconductor industry as a whole (which was subsidized by the government in its early years for aerospace applications, some civilian and some not).
But that said it's more about industrial capacity than research. We build planes and tanks not because we need them today, but because we might need the ability to build them tomorrow. Continued research on military applications prevents institutional knowledge from dying and tools from rotting.
>> They complement one another too.
>> One helps the other. The reverse direction is slower though because it is often held close to the vest for a bit of time.
Glad we agree
> We build planes and tanks not because we need them today, but because we might need the ability to build them tomorrow.
I also agree. I'm not sure why people think I am saying to not build tanks and planes during peace time. I just think we should be funding civilian science more and that the military benefits from such research. I haven't really disagreed with all these people "disagreeing" with me.
Consider the US Air-force is unquestionably the #1 most powerful Air-force in the world with the US Navy comfortably holding the #2. That’s frankly expensive.
Unfortunately, many NATO countries are currently spending far less than 2% (as you showed above)... and the US is spending far more than 2%, in part, to make up for the shortfall (ie, why we have tens of thousands of troops stationed in Europe, etc).
[1]: https://foreignpolicy.com/2016/02/03/if-russia-started-a-war...
Any nation in history will lose a single battle if their enemies concentrate on a single point and their defending a huge area. However, such a loss represents no significant reduction in fighting forces and is therefore not militarily meaningful. Further a massive defensive concentration of forces into a small area just means the actual attack takes place somewhere else.
In the end winning requires armies to face armies or supplies to run out over long periods.
That's a major change. During the Cold War, absent tactical nukes, NATO loses continental Europe to the Warsaw Pact.
You dont need to be a direct military competitor, Afghanistan and Iraq have shown that you can still defeat US with a thousand cuts.
George Bush in 2003 had the Mission Accomplished banner, if the mission was accomplished, what are US soldiers still doing in Afghanistan and Iraq.
Which means the US had to do all the expensive logistics work too.
At the expense of US soldiers dying from the hands of AL Qaeda/ISIS and wasting trillions of dollars which could have been used to rebuild US infra and providing healthcare and other services to US citizens.
Didn't the US lose by that definition?
No one is saying US does not have that monopoly, its just that US still loses with that monopoly.
You could just as well double the EU's relative share by halving the US military budget. The US would still spend more than China, Russia, and a bunch of other countries combined. Surely that's more than enough.
It's just that US politicians always frame it as if the total spending had.to remain constant.
Navy said, great, let's turn it into a tanker!
Northrop said nope but Boeing said they'd be happy to. So billions as long as those robots stay in their lane.
The Pentagon has been using robots to kill robots and other humans since before 9/11. This story has a long arc. We meat bags are the major limitation for machines of war and that limitation will be engineered out.
And regardless, is there even value in having autonomous fighter jets when we already have advanced missiles? It doesn't matter how good the pilot/AI is if they are just firing missiles from miles away. We already have missiles that can loiter for hours before choosing a target and attacking it (ex. IAI Harpy) - to me that is much more valuable and dangerous than an AI fighter.
Army of None: Autonomous Weapons and the Future of War is a good book to read about it.
https://en.wikipedia.org/wiki/A_Taste_of_Armageddon
> In the episode, the crew of the Enterprise visits a planet engaged in a completely computer-simulated war with a neighboring planet but the casualties, including the Enterprise's crew, are supposed to be real.
It's one thing to go a mile outside your airspace to nab somebody's drone. To do that to nab someone else's pilot is a whole 'nother level of escalation.
Actually, the US military prepared multiple counter strike options, and the president (eventually) picked one.
He then called off the strike a few hours before the time on target, because a faction in the Whitehouse gave him an estimate that the counterstrike would result in 150 casualties.
See John Bolton's book "The Room Where It Happened" for a very elaborate inside account. He is very open that he would've preferred the strike to go through, and while he believes above casualty estimates were much too high, he was very much aware the number would not be zero.
I can see how any US casualties might've tipped the balance in Trump's mind.
So yes, there might be some value. But where it gets complicated is whether the value is positive or negative:
One could say that the US flying only an autononous aircraft that day saved up to 151 lifes.
The famous example is the US’ 10k troops in Berlin during the Cold War. If the USSR attacked, those troops were toast - and everyone knew it. But it meant that if the USSR took Berlin, it would be at the price of 10k American soldiers, and would inevitably lead to war. Even if that was the US’ calculation anyway, the Soviets wouldn’t know that for certain without the troops there.
When the guy who proposed the idea, Roger Fisher, was asked who the hell would ever volunteer to be the nuclear pinata, he volunteered to do it himself.
https://en.wikipedia.org/wiki/Roger_Fisher_(academic)#Preven...
This is probably the important bit, and the bit that most needs a human in the loop. Once the decision is made, engage the AI and let it do its thing.
Of course, if some regime out there didn't care at all about ROE or human life or killing the wrong people, they could build a bunch of killing machines to take out anything that moves. It would be more efficient.
That's a good take on the matter. Computer systems are already a big part of the aircraft, in the near term this could result in some pretty wicked copilot software.
(I don't know what the search terms would be in chinese[1], but "искусственный интеллект су-35" gets plenty of older —at least by HN standards— hits.)
[1] https://www.bbc.com/news/world-asia-china-28905504 suggests china has bought while they learn to build. Someone who knows more chinese than I might wish to try "人工智能 苏35".
Because it was a simulation, the human didn't black out. Its just a game. Albeit a well simulated game close to dogfighting "reality", but its a game nonetheless.
An intermediate step could be an AI-assisted aircraft. It might be better at setting up shots, and still work within the physical limitations of the human inside.
Something not mentioned is an AI powered plane's ability to more-rapidly change from positive to negative G's (not sure if this is technically called jerk or not). A human has a really tough time making that transition.
An unmanned aircraft could laughably outmaneuver an F-22. The AI is the tricky part.
Dogfighters see one another pull up (or dive initiate) and expect a loop, and so they think in maneuver sequences that occur in 5-10 second chunks. Also a cadence of countermeasure ejection turn-taking to deplete opponent heat-seeking payload, punctuated in between fixed projectile fire. "Checkmate" occurs when a dogfighter has successfully "landed" a loop or sequence of maneuvers that gives them a shot behind their opponent.
I'm wondering if the modern combat aircraft would have gimballed barrels and auto-aiming (non-projectile speed of light high intensity laser could melt a fuel tank). If this were true, then by default pilots are already setting up shots as you say using a form of AI assistance.
The question I have is, auto-aiming aside, how would an NN understand how to coordinate these "chess moves". This question becomes irrelevant if suddenly there could be a positive to negative G change mid-loop that no seasoned dogfighter could ever expect to happen.
Current fighters wear out after so many flight hours. What puts the majority of these hours on is pilot training. With an AI commanded aircraft, there's little need to actually fly it, other than to demonstrate capability occasionally so that potential opponents see that you have a credible offense. So you build them to essentially be disposable - rather than design for a 10,000 hour lifetime, you design for a 500 hour (rather intense!) lifetime.
The HUD is cool and all. But how do I know where the enemy is if it is outside of the HUD's field? Think about how the AI knew. :) And while we are at this, maybe just invite a young gamer instead of a real pilot.
(Oh, and I fail to see how China wouldn't win. Heh.)
That just leaves "counter-insurgency" empire building crap which is super inefficient at even the stated goals, and the stated goals are worthless.
Therefore I conclude the whole exercise is pointless and a waste of money.
In 2015.
There is nothing special about China and Russia having similar systems.
The same military requirements apply everywhere.
And just because in the US it is not be popularized as "killer robots" does not mean that it is not already that.
Its kind of silly that some people think that the US military has some type of military moral high ground in this case or any. Or that by not explicitly saying "killer robots" that means it is not so.
If the engineers in charge of designing these crafts around an AI pilot think outside the box we could see some pretty wild tech.
Those maneuvers sound like they'd destroy the plane.
I'm not sure if that would work. Missiles can put a lot more stress on the airframe because they're literally single-use disposable items. Something that takes the "plane" role would probably be larger (so missile solutions may not scale) and would have to account for things like fatigue over time.
I'm reminded of the Mig 25: the planes are capable of flying at Mach 3.2, but doing so wreaks the engines. If it was a missile, it'd operate a full speed no question; but since it's a reusable plane, it's operationally limited to Mach 2.8.
Well, we can already do things like this with non-piloted tech; and there is no way that a traditional jet could keep up, load-wise. The g-force induced load limit for most modern jets is about double the rated limit. You likely wouldn't be able to fly any air-frames made with human passengers in mind over 35g without significant re-engineering.
The point of the AI jet-fighter idea is to have an AI that is capable of piloting a fighter jet (designed for a human) at near-human levels.
We can already programmatically control super fast flying devices into whatever wacky flight-plan we calculate on the fly, on the device, near instantly. Lockheed and Raytheon advertisements have been talking about super-fast on-device radar/surveillance avoidance systems for years.
> we could see some pretty wild tech.
you're not kiddin', the only thing that outweighs my hope for what the future brings is my fear of what the future brings.
If modern fighters are missile platforms, why not design a fleet of cheap no-human drone missile platforms and put most of the intelligence into the missiles?
The "wild tech" would mean getting rid of super-expensive low-reliability fighters altogether, because it's not obvious what they bring to an automated conflict.
I wonder why we're even talking about planes any more. Missiles to kill the planes and missiles to attack the ground. No pilots, no problems. Right?
> Because we need a human up there, flying around, controlling the missiles
Do we? Why is this important. I thought everything could be done remotely these days.
Jamming is an issue for one. Latency for another.
IE an "AI aircraft" is a smart missle that can come home.
In terms of tactical decision making while flying, I think AI's advantage is its ability to integrate (much) more data into overall situational awareness. If you were somehow able to integrate all of this directly into the human proprioception system I have a feeling we would fare much better. It seems like we're probably quite a bit further from humans being defeated in contact sports by an AI-powered robot of equivalent strength/weight/mobility.
Not that much further though. :/
The reality is that aircraft have to fly in air, which imposes severe physical limits regardless of whether there is a human pilot on board.
If I were designing such an aircraft I would make sure that the onboard missiles could be rapidly swiveled. Someone on your tail? Easy, just swivel a missile around 180 degrees and then fire. No need to maneuver the entire aircraft.
A swiveling missile launcher is far from easy. Please calculate how much such a contraption would weigh. Remember that it needs to securely hold multiple large, heavy missiles without structural deformation during your hypothetical 10G maneuvers. And what sensors will you use to detect someone on your tail and target your own missile before he does? Where will those sensors be placed, how much will they weight, and how much drag will they add?
The concepts you propose only work in cartoons, not in the real world bound by laws of physics.
30G is not within reach. Calculate how much the airframe would have to weigh to handle that strain, and explain how to avoid stalling the wing or compressor. Scaling up a small missile design won't work at all. This is just basic physics. Range and radar signature are higher priorities than maneuverability.
https://www.npr.org/2020/08/17/903152237/bald-eagle-in-michi...
I don't think he's saying that at all. On the contrary, he's saying the opposite, explaining the advantages an AI has in these areas. For example,
> Even proficient aviators have to use a percentage of their concentration (i.e. situation awareness) on not over-performing or under-performing the aircraft. AI could easily track this task and would most likely never bleed airspeed or altitude excessively, preserving vital potential and kinetic energy while also fine-tuning lift vector placement on the other aircraft to continue the fight if required.