A Grandfather* was tossed 8 FEET into the Air by a Bison! Film at 11 !!!
Grumpy
- You would think that somebody who has lived that long to become a Grandfather. Would be old enough to know better!
Category: Paint me surprised by this
Artificial intelligence (AI) isn’t good at rendering firearms, and that’s probably not a bad thing, for reasons that will soon be explored. It should also be noted that while AI rendered guns isn’t good yet, it is likely to improve. That might not be a good thing either.
First, we need some background on what exactly AI is, or more accurately, what is generative AI. It has been in development for decades, but it was only in 2022 that generative AI entered the public consciousness with consumer-friendly programs.
Suffice it to say that generative AI, which includes OpenAI’s ChatGPT, Anthropic’s Claude, and Google Gemini, among many others, is the evolution of technology that emerged in the 1960s and is something that is now used every day, again both for good and bad.
That’s not just in how it renders firearms.
AI can create visual artwork, compose music, write papers and stories, and produce other content. Students are using it as a shortcut with homework, and anyone who has been on social media has likely seen images and videos of varying quality. With a few prompts, which are written instructions, AI can generate content that just a few years ago would have taken design teams days, even weeks to produce.
However, there is a lot of “AI slop,” as in low-quality, mass-produced digital content that shows a lack of human effort. Such content is seemingly created — if created is really the right term — by those seeking to flood the Internet to exploit algorithms and to generate quick ad revenue on some platforms.
How AI Works
It is important to understand how AI images are generated. How “intelligent” AI actually is remains a matter of debate.
Generative AI is based on “Large Language Models” (LLMs), which are trained on massive datasets of text and code to recognize patterns. Although they can be effective at parsing data, solving logic puzzles and even summarizing lengthy documents, LLMs can also produce factually incorrect, nonsensical, or made-up information that may seem plausible.
AI generates images by “learning” patterns from billions of existing pictures and their accompanying text descriptions. This process is called diffusion. AI is very good at generating images of well-known objects, where the attributes, including dimensions, color, and other properties, are clearly described.
It would seem that AI, therefore, should be good at rendering firearms, but a search of social media tells a very different story. AI-generated images of firearms depict weapons that would be impossible to produce in the real world, much like the fantastical settings created by the late Dutch artist M.C. Escher.
In the case of firearms, LLMs and AI have no shortage of information to draw from, yet generative AI platforms still struggle because they lack a true 3D understanding of the underlying mechanics, resulting in nonsensical attachments, bent barrels, missing triggers, and incorrect magazine placement. AI is still treating firearms as abstract collections of shapes rather than focusing on the functional mechanisms. That has resulted in impossible geometry.
There are now Facebook Groups and Reddit subs devoted to sharing AI-generated firearm images that defy reality. Even when it gets the basics right, AI still misses some key details. Several factors are at play, but the most basic is that generate AI is based on algorithms, and many are far less “intelligent” than it might seem.
Guns And Media — It’s Never Been Accurate
It is further worth taking a step back and remembering that mainstream depictions of firearms have long been questionable. Consider firearms in comic books and video games as just two examples.
Writers of the former and developers of the latter would routinely ignore basic mechanics, notably magazine capacities and recoil, opting instead to treat firearms as versatile plot devices that are only as accurate or deadly as the story demands at any given moment.
Likewise, the World War II-based comic books of the 1960s didn’t bother to feature realistic depictions of the enemy. The Germans were often presented with big red swastikas on their helmets, carrying weapons that were not an accurate drawing of the MP-40. It was a version that fit the narrative, even if it was far from accurate.
Video games have, in recent years, gone to great lengths to get many details of modern firearms right, including the look and sound, yet other attributes are often still very wrong, notably the weight and recoil. The guns may look correct, but they way they are employed and operate is anything but accurate.
Lack of Instruction?
AI continues to struggle with firearms for some very simple reasons. There is an old saying among computer programmers: “Garbage in, garbage out,” which is the principle that the quality of a system’s output is directly linked to the quality of input. Flawed or low-quality data will produce equally flawed or useless results.
In the case of AI, the “garbage in” is the lack of clear instructions.
“One of the reasons that guns are not properly rendered in AI is in the details,” explained Roger Entner, founder and principal analyst at Recon Analytics. “AI is only as good as the instructions you give it.”
AI often needs more information and details than it is given. It can get the basic shapes right, but it still doesn’t understand the mechanics.
“Guns are such intricate tools, with minuscule differences that are huge,” Entner told The Armory Life. “Gun owners may know these things so intricately, but AI does not.”
That is a key point to consider. There is already massive confusion in the mainstream consciousness about the differences between a commercial AR-15 and the military M16, so how can we expect AI to know better?
AI doesn’t just fall short with firearms, but all sorts of things.
“AI isn’t good at depicting a Cadillac XTS from a Chevrolet Malibu,” said Entner.
There is also the issue of bias, and it is firearm enthusiasts who are likely to notice the rendering problems of AI when it comes to guns.
“AI really does render firearms poorly. Many details range from implausible to outright wrong. But in perspective, AI renders many objects poorly,” noted Dr. Jim Purtilo, associate professor of computer science at the University of Maryland.
“If someone in the firearm community notices it with guns more than with other objects, then this could well be another example of a selective attention bias, which is where people with a decided interest in a given object will be much more inclined to notice when details are wrong. Show a young man the photo of a pretty young girl, and he will never notice what is in the background. ‘What monster?’”
Such a selective attention bias has cropped up in cinema for years.
“Show an old war movie and the gun enthusiast will complain, ‘that’s not a rifle they used,’ the history aficionado will say, ‘that’s not how it happened,’ and the linguists will lament that characters used words that didn’t become common until the modern era. This illustrates our biases,” Purtilo told The Armory Life.
Still, the implausible images result from how AI generates them today. There is ample data on firearm attributes, but few descriptions of how everything connects. AI’s LLMs make a best guess, often with comedic results.
“Pictures are tougher for AI than we understand,” added Entner. “Generic descriptions of something to AI will generate a generic description. It is like asking a five-year-old to draw a Single Action Colt.”
AI is thus like a child, and it may not provide all the details unless pressed. Even then, it may not fully appreciate how things go together without further explanation.
“Said simply, the program is averaging all the details of its training images when deciding which features to include,” said Purtilo. “It might know that a firearm has sights, a shoulder stock, and attachments, but it doesn’t know how these details might depend on one another.
The average sight across all the images it analyzed might have been a scope, the average stock might have a pistol grip, and the average attachment might be a laser pointer — and that is how it gives you a Revolutionary War musket with pistol grip, high power optic, and laser pointer.”
The Barriers of AI
The current technical barriers are just one of the main reasons that AI is bad at rendering firearms. The other is policy. AI developers are already cautious about how AI can be used.
“On the technical side, guns are complex objects with very specific features. Because they are so specific and have many parts that can be rendered incorrectly, it’s easy for a human to identify them as incorrect, as AI is not good at replicating specific weapons,” said Dr. Cliff Lampe, professor of information and associate dean for academic affairs in the School of Information at the University of Michigan.
Lampe, who focuses on the study of misinformation in media, told The Armory Life that the policy reason is that many models specifically list firearms as a type of object to render poorly.
“You can imagine, for instance, that you don’t want to be able to have GenAI [generative artificial intelligence] create specific schematics for firearms. Different models may have different thresholds here as a matter of policy, but all of them will have some safety restrictions built in,” Lampe noted.
Microsoft’s Copilot and ChatGPT are now among the generative AI platforms that won’t even render a firearm if requested. Copilot won’t even render soldiers holding firearms. However, because of the rise in “AI slop,” it may be a good thing that AI can’t generate extremely accurate firearms in videos. No manufacturer would be happy to see their product — a car, gun, or something else — used irresponsibly in an AI-generated video.
For now, we may need to accept that AI doesn’t do guns well, just as comic writers and video game developers missed the mark in the past.
“I don’t know whether AI is more likely to generate silliness with firearms than other objects. But if so, then this may well reflect a selection bias in training materials,” Purtilo continued. “AI models are trained by ingesting a huge volume of raw content. The companies scrape pages from websites, books, social media, and more for this purpose.”
As Lampe noted, there is also the issue of policies that restrict many sites from including firearm images, which could impact how AI learns.
“This limits what material AI has for training,” said Purtilo. “If you can’t depict safe and responsible firearm use on the web, then no AI will be able to render images with safe and responsible use of firearms. Whatever it generates will be inherently gibberish.”
———————————————————————————-
Wolf the Husky asked ““How in the heck did you do that from all the way over there?”
We know-it-all outdoor writers like to project an aura of absolute competence. If you read enough hunting and firearms literature, you’ll start to believe every writer lives in a pristine cabin tastefully decorated with big game trophies, a rack of flawless antique shotguns and a kennel full of immaculate, perfectly disciplined bird dogs that obey whispered commands and hand signals from a mile away.
Allow me to shatter that illusion.
This is what all pet owners dream of when they buy a cute little puppy.
Long before I was writing about firearms full-time, I was just another desperate hunter locked in a daily battle of wits against two lawless canine insurgents. Looking back at those early days as a dog owner, I am reminded that the relationship between a man, his canines and outdoor technology has always been a rocky one.
In fact, looking back at the journals from my early days reminded me of just how ugly the battle for supremacy inside the fence line can get.
I have always been deeply envious of families who own real dogs. You know the type: intelligent, faithful, fiercely loyal, obedient creatures content to sit lovingly at your feet beside a roaring fire on a cold winter night. Our friends had a dog named Molly, like that, who could probably solve quadratic equations and instinctively knew which fork to use for the salad course.
Instead, we had Jake The Wonder Dog and Wolf.
If dogs could talk, you know they’d always have some kind of elaborate explanation about how the mess wasn’t his fault, and he’s really, really sorry, and he’ll never do it again … until next week.
Rogues Gallery
To be fair, our dogs fulfilled their primary missions reasonably well. Jake, a veteran retriever, was responsible for locating downed birds and bringing them back to the blind. However, he was rapidly reaching the age where he’d look at me after a shot with an expression that clearly said, “You shot it, pal. You go get it.”
Wolf, an Alaskan Husky and the newest member of the crew, was brought on board to assist in those moments when one dog wasn’t generating enough household pandemonium. Wolf was a fine specimen of young male doghood, and under Jake’s careful corrupting, he was also rapidly becoming completely unmanageable.
Our problems stemmed from the fact both dogs were entirely lawless. While they occasionally submitted to the absolute will of Yours Truly — Master of the Realm and Keeper of the Kibble — they were equally happy to flatten smaller family members while running pell-mell toward the horizon. Their high-speed flights were usually only halted when they proved too stupid to avoid the local dogcatcher.
In this regard, Jake, if he were human, would be known as a “repeat offender” and likely to get 30 years hard time on his next encounter with the law.
Combined with their natural talents for excessive barking, generalized kennel destruction, and a collective personality that would be described as “busy” if they were human children, they weren’t really part of the family — they were an occupying force.
We searched desperately for answers to the myriad behavioral problems. We bought books, watched videos, sought out high-priced experts and spent more time in obedience training than we did with our actual human children. The end result was always the same: each evening, our neighbors would gather on their porches to enjoy the chaotic, gladiatorial spectacle of daily feeding time.
One expert consultant summed up our situation best: “If I were you, I’d move and not tell the dogs.”
Modern Miracle
Everything changed when we finally decided to purchase that modern miracle of animal training: the invisible fence.
The concept behind this device is simple. A wire buried along the perimeter of your yard sends out a radio signal detected by a special collar worn by the dog.
As the dog approaches the boundary, he is warned by a series of increasingly frantic beeps. If he continues to ignore the warning, he is given what the owner’s manual calls “a mild electrical correction.” Over time, the dogs are supposed to recognize the boundary markers and stay within the safe zone.
I was highly skeptical and deeply concerned that this fairly expensive system would be entirely useless against Jake, a dog known to gnaw through solid concrete out of sheer boredom. I fully assumed he would walk right up to the boundary, take the shock, and look back at me like, “Huh. Is that the best you got, Chief?”
How wrong I was.
After showing the dogs the new boundaries, I unclipped their leashes and released them into the yard. The reaction from each beast was fascinating from a behavioral standpoint — and, frankly, hilarious from the slightly sadistic viewpoint of a man who had spent years chasing them through the muddy fields in the middle of the night, constantly apologizing to the neighbors.
Flight and Fallout
Wolf was a happy-go-lucky soul with zero pretensions of being an Alpha male. I caught him just as he attempted to sneak next door to visit his new girlfriend. Channeling my inner police officer, I bellowed, “STOP!”
True to form, Wolf glanced over his shoulder then, feigning sudden deafness, took another step. The results were immediate and dramatic.
When the collar issued its “mild correction,” Wolf launched straight into the air like a Polaris missile, executed several impressive mid-air twists and attempted to gain altitude as he streaked back to the absolute safety of his doghouse.
Once inside, he stuck just his nose and one suspicious eye out of the doorway. The bewildered look on his face made his thoughts entirely transparent: “How in the heck did you do that from all the way over there?”
Jake was a completely different story. Being the stubborn old hand, he confidently strode up to the line, gave me a clear dismissive look and boldly stepped across.
I will omit the graphic details of his immediate physical reaction, except to say that it created quite a bit of distasteful cleanup work. On the bright side, the grass in that particular section of the yard was much greener for the rest of the season.
Jake the Wonder Dog was eventually sentenced under Indiana Criminal Code Title 9 to spend 15 hunting seasons retrieving downed waterfowl, even when the water was really, really cold.
Peace Dividend
Within a week, the transformation in our dogs was nothing short of miraculous. Allowed to roam the yard free from tethers, leashes and kennels, they became much happier, calmer and genuinely fun for our kids to be around. For the first time in years, we became that picturesque, happy family with pets lying contentedly at our feet as the sun went down.
While I don’t universally advocate for electronic training collars due to their high potential for misuse by irresponsible owners, the invisible-fencing concept certainly saved my two outlaws from a one-way trip to the county pound — or worse.
Besides, I can tell you from personal experience that the shock really isn’t that bad.
I discovered this firsthand while carrying both collars in my left hand during a walk from the garage to the backyard. I am happy to report that I eventually regained almost full use of that arm.
Modern Takeaway
Decades have passed since Jake and Wolf ran the perimeter of my yard, and I have officially outlived every single four-legged culprit from that chaotic era of my life.
Today, we have satellite-guided GPS collars which have microprocessors that can track a dog’s heart rate from three counties away, and selectable shock levels ranging from a polite tap on the shoulder to a setting known as “Horse Defibrillator.” The gear has evolved, becoming safer, smarter, and infinitely more humane than what I had to work with.
But as I look at the sleek, high-tech training systems on the market today, I smile because no matter how advanced the microchips get, they still have to contend with the oldest, most stubborn force on the planet: a bird dog who thinks he knows better than you do.
Then again, at least he didn’t try to carry two shock collars out of the garage.
He led a reconnaissance team during a mission near Phú Lộc, Vietnam, in 1967. During the engagement, Capers sustained multiple wounds while directing his men and coordinating their evacuation under intense enemy fire. His actions were instrumental in saving the lives of every member of his team.
____________________________________________________________
Pity it took so long to get it to him! (Almost 60 years!!) But that’s the problem with the Blue Max. In that sometimes you get it ASAP & other times it’s almost too late. I say that it’s just pure BS and a few officers need a blow torch pointed at them. Grumpy
In late May Chinese leaders travelled to the Zhoushan National Oil Reserve and discovered the nation’s strategic oil reserves weren’t there. For over a year, the disruption of oil supplies from Venezuela and Iran had left Chinese oil reserves reduced. Despite that, government documents indicated that China still had 1.2 billion tons of oil reserves. That’s equivalent to 8,756,117,022 barrels.
China’s strategic oil reserve, to the surprise of the government officials who went to verify the reserves in May, was instead composed of water, sludge, various debris and overflow from nearby sewer lines.
Because the Americans dominated global energy supplies, the Chinese oil reserve served as a major cushion to any disruptions to Chinese oil imports from the Persian Gulf, especially Iran whose main customer was China. Under America’s global energy stranglehold, Chinese crude oil stockpiles have reached the verge of collapse at the slightest exposure.
The current Chinese vulnerability stems from the American disruption of Venezuelan oil exports to China and more recently a similar situation with Iranian oil exports to China.
China’s strategic oil reserve was insurance against disruptions in Venezuelan and Iranian imports. With its oil reserves revealed as a sham, China finds itself in a desperate situation.
What happened to Chinese oil? It was soon discovered that corrupt government officials and oil reserve personnel had sold the oil and pocketed the proceeds. The local buyers were often operators of small, locally owned refineries that turned the oil into commercial products that were sold throughout China.
Most of these oil criminals then fled, often leaving China for sanctuary states that would welcome any affluent Chinese and their new wealth. The only winners were a few conniving Chinese and the Americans, who continued to dominate the global energy system.
In China corruption is not just an economic issue. For thousands of years there was corruption in the military as well. China has had a corruption problem in its military for a long time. Think thousands of years. One reason the communists won the civil war against the Nationalists in the 1940s was because the Nationalists were much more corrupt.
For about a generation, the communists kept corruption under control. But for the last four decades, corruption has been a growing problem. This despite several major efforts to stop it.
The theft of China’s strategic oil reserves is only the latest of several recent awful discoveries of massive corruption impeding the ambition of its leader, President Xi, to conquer Taiwan.
The last one concerned military corruption in building equipment to invade Taiwan, and was discovered in 2023. Missile fuel tanks were found to be filled with water, missile silo lids could not be opened, and the protective concrete missile silos themselves were so defective they might as well have been made of wood.
The air force and navy lacked sufficient spare parts for even a week of operations against Taiwan and many aircraft and naval vessels were outright inoperable. This postponed the earliest possible date for the invasion of Taiwan from 2025 to 2027.
The non-existence of China’s strategic oil reserve will probably have the same effect. The on-going closure of the Strait of Hormuz due to the war with Iran means China cannot begin to refill its oil reserve until several months after that war ends, due to the need to clear the Strait of anti-ship mines.
Then it will take at least another 18 months for China to rebuild, at exorbitant expense to its hard currency reserves, the 1.2 billion ton oil reserve it thought it had. China’s natural gas reserves, to the extent those were believed to exist, were quite inadequate to its needs in the event of a US blockade in a war with Taiwan. Closure of the Strait of Hormuz has made that clear. It will take still longer and cost more to build a strategic natural gas reserve.
There are at least major possibilities that a Chinese invasion of Taiwan has been postponed for another two years, to 2029, and perhaps indefinitely due to further discoveries of calamitous-scale corruption.
So now the Chinese are really getting serious. They are installing cost control systems, with regular audits, to monitor their military spending. In the past, budgeting was pretty primitive, mainly because cost accounting was expensive to implement.
In the past, detailed spending data was collected after the fact, if at all. This made it easier for corrupt officers to steal. Noting how effectively Western, Japanese and Taiwanese companies dealt with corruption using detailed budgets and frequent audits, China installed similar systems throughout its armed forces.
This caused some morale problems among senior officers, but this was not believed to be serious. Meanwhile, the program cut costs an average of ten percent in units where it had been installed.