Artificial Intelligence is a once-in-a lifetime commercial and defense game changer

(download a PDF of this vendible here)

Hundreds of billions in public and private wanted is stuff invested in Strained Intelligence (AI) and Machine Learning companies. The number of patents filed in 2021 is increasingly than 30 times higher than in 2015 as companies and countries wideness the world have realized that AI and Machine Learning will be a major disruptor and potentially transpiration the wastefulness of military power.

Until recently, the hype exceeded reality. Today, however, advances in AI in several important areas (here, here, here, here and here) equal and plane surpass human capabilities.

If you haven't paid attention, now's the time.

Artificial Intelligence and the Department of Defense (DoD)
The Department of Defense has thought that Strained Intelligence is such a foundational set of technologies that they started a defended organization- the JAIC – to enable and implement strained intelligence wideness the Department. They provide the infrastructure, tools, and technical expertise for DoD users to successfully build and deploy their AI-accelerated projects.

Some specific defense related AI applications are listed later in this document.

We're in the Middle of a Revolution
Imagine it's 1950, and you're a visitor who traveled when in time from today. Your job is to explain the impact computers will have on business, defense and society to people who are using transmission calculators and slide rules. You succeed in inveigling one visitor and a government to prefer computers and learn to lawmaking much faster than their competitors /adversaries. And they icon out how they could digitally enable their merchantry - supply chain, consumer interactions, etc. Think well-nigh the competitive whet they'd have by today in merchantry or as a nation. They'd steamroll everyone.

That's where we are today with Strained Intelligence and Machine Learning. These technologies will transform businesses and government agencies. Today, 100s of billions of dollars in private wanted have been invested in 1,000s of AI startups. The U.S. Department of Defense has created a defended organization to ensure its deployment.

But What Is It?
Compared to the archetype computing we've had for the last 75 years, AI has led to new types of applications, e.g. facial recognition; new types of algorithms, e.g. machine learning; new types of computer architectures, e.g. neural nets; new hardware, e.g. GPUs; new types of software developers, e.g. data scientists; all under the overarching theme of strained intelligence. The sum of these feels like buzzword bingo. But they herald a sea transpiration in what computers are capable of doing, how they do it, and what hardware and software is needed to do it.

This unenduring will struggle to describe all of it.

New Words to Pinpoint Old Things
One of the reasons the world of AI/ML is troublemaking is that it's created its own language and vocabulary. It uses new words to pinpoint programming steps, job descriptions, minutiae tools, etc. But once you understand how the new world maps onto the archetype computing world, it starts to make sense. So first a short list of some key definitions.

AI/ML – a shorthand for Strained Intelligence/Machine Learning

Artificial Intelligence (AI) – a catchall term used to describe "Intelligent machines" which can solve problems, make/suggest decisions and perform tasks that have traditionally required humans to do. AI is not a single thing, but a constellation of variegated technologies.

Machine Learning (ML) – a subfield of strained intelligence. Humans combine data with algorithms (see here for a list) to train a model using that data. This trained model can then make predications on new data (is this picture a cat, a dog or a person?) or decision-making processes (like understanding text and images) without stuff explicitly programmed to do so.

Machine learning algorithms – computer programs that retread themselves to perform largest as they are exposed to increasingly data. The "learning" part of machine learning ways these programs transpiration how they process data over time. In other words, a machine-learning algorithm can retread its own settings, given feedback on its previous performance in making predictions well-nigh a hodgepodge of data (images, text, etc.).

Deep Learning/Neural Nets - a subfield of machine learning. Neural networks make up the windrow of deep learning. (The "deep" in deep learning refers to the depth of layers in a neural network.) Neural nets are constructive at a variety of tasks (e.g., image classification, speech recognition). A deep learning neural net algorithm is given massive volumes of data, and a task to perform – such as classification. The resulting model is capable of solving ramified tasks such as recognizing objects within an image and translating speech in real time. In reality, the neural net is a logical concept that gets mapped onto a physical set of specialized processors. See here.)

Data Science - a new field of computer science. Broadly it encompasses data systems and processes aimed at maintaining data sets and deriving meaning out of them. In the context of AI, it's the practice of people who are doing machine learning.

Data Scientists – responsible for extracting insights that help businesses make decisions. They explore and unriddle data using machine learning platforms to create models well-nigh customers, processes, risks, or whatever they're trying to predict.

What's Different? Why is Machine Learning Possible Now?
To understand why AI/Machine Learning can do these things, let's compare them to computers surpassing AI came on the scene. (Warning - simplified examples below.)

Classic Computers

For the last 75 years computers (we'll undeniability these classic computers) have both shrunk to pocket size (iPhones) and grown to the size of warehouses (cloud data centers), yet they all unfurled to operate substantially the same way.

Classic Computers – Programming
Classic computers are designed to do anything a human explicitly tells them to do. People (programmers) write software lawmaking (programming) to develop applications, thinking a priori well-nigh all the rules, logic and knowledge that need to be built in to an using so that it can unhook a specific result. These rules are explicitly coded into a program using a software language (Python, JavaScript, C#, Rust, ¦).

Classic Computers – Compiling
The lawmaking is then compiled using software to translate the programmer's source lawmaking into a version that can be run on a target computer/browser/phone. For most of today's programs, the computer used to develop and compile the lawmaking does not have to be that much faster than the one that will run it.

Classic Computers – Running/Executing Programs
Once a program is coded and compiled, it can be deployed and run (executed) on a desktop computer, phone, in a browser window, a data part-way cluster, in special hardware, etc. Programs/applications can be games, social media, office applications, missile guidance systems, bitcoin mining, or plane operating systems e.g. Linux, Windows, IOS. These programs run on the same type of archetype computer architectures they were programmed in.

Classic Computers - Software Updates, New Features
For programs written for archetype computers, software developers receive bug reports, monitor for security breaches, and send out regular software updates that fix bugs, increase performance and at times add new features.

Classic Computers- Hardware
The CPUs (Central Processing Units) that write and run these Archetype Computer applications all have the same vital diamond (architecture). The CPUs are designed to handle a wide range of tasks quickly in a serial fashion. These CPUs range from Intel X86 chips, and the ARM cores on Apple M1 SoC, to the z15 in IBM mainframes.

Machine Learning

In unrelatedness to programming on archetype computing with stock-still rules, machine learning is just like it sounds - we can train/teach a computer to "learn by example" by feeding it lots and lots of examples. (For images a rule of thumb is that a machine learning algorithm needs at least 5,000 labeled examples of each category in order to produce an AI model with decent performance.) Once it is trained, the computer runs on its own and can make predictions and/or ramified decisions.

Just as traditional programming has three steps – first coding a program, next compiling it and then running it – machine learning moreover has three steps: training (teaching), pruning and inference (predicting by itself.)

Machine Learning – Training
Unlike programing archetype computers with explicit rules, training is the process of "teaching" a computer to perform a task e.g. recognize faces, signals, understand text, etc. (Now you know why you’re asked to click on images of traffic lights, navigate walks, stop signs, and buses or type the text of scanned image in ReCaptcha.) Humans provide massive volumes of "training data" (the increasingly data, the largest the model's performance) and select the towardly algorithm to find the weightier optimized outcome. (See the detailed "machine learning pipeline" section for the gory details.)

By running an algorithm selected by a data scientist on a set of training data, the Machine Learning system generates the rules embedded in a trained model. The system learns from examples (training data), rather than stuff explicitly programmed. (See the "Types of Machine Learning" section for increasingly detail.) This self-correction is pretty cool. An input to a neural net results in a guess well-nigh what that input is. The neural net then takes its guess and compares it to a ground-truth well-nigh the data, powerfully asking an expert "Did I get this right?" The difference between the network's guess and the ground truth is its error. The network measures that error, and walks the error when over its model, adjusting weights to the extent that they unsalaried to the error.)

Just to make the point again: The algorithms combined with the training data – not external human computer programmers – create the rules that the AI uses. The resulting model is capable of solving ramified tasks such as recognizing objects it's never seen before, translating text or speech, or executive a drone swarm.

(Instead of towers a model from scratch you can now buy, for worldwide machine learning tasks, pretrained models from others and here, much like tweedle designers ownership IP Cores.)

Machine Learning Training – Hardware
Training a machine learning model is a very computationally intensive task. AI hardware must be worldly-wise to perform thousands of multiplications and additions in a mathematical process tabbed matrix multiplication. It requires specialized fries to run fast. (See the AI semiconductor section for details.)

Machine Learning – Simplification via pruning, quantization, distillation
Just like archetype computer lawmaking needs to be compiled and optimized surpassing it is deployed on its target hardware, the machine learning models are simplified and modified (pruned) to use less computing power, energy, and memory surpassing they're deployed to run on their hardware.

Machine Learning - Inference Phase
Once the system has been trained it can be copied to other devices and run. And the computing hardware can now make inferences (predictions) on new data that the model has never seen before.

Inference can plane occur locally on whet devices where physical devices meet the digital world (routers, sensors, IOT devices), tropical to the source of where the data is generated. This reduces network bandwidth issues and eliminates latency issues.

Machine Learning Inference – Hardware
Inference (running the model) requires substantially less compute power than training. But inference moreover benefits from specialized AI chips. (See the AI semiconductor section for details.)

Machine Learning - Performance Monitoring and Retraining
Just like archetype computers where software developers do regular software updates to fix bugs and increase performance and add features, machine learning models moreover need to be updated regularly by subtracting new data to the old training pipelines and running them again. Why?

Over time machine learning models get stale. Their real-world performance often degrades over time if they are not updated regularly with new training data that matches the waffly state of the world. The models need to be monitored and retrained regularly for data and/or concept drift, harmful predictions, performance drops, etc. To stay up to date, the models need to re-learn the patterns by looking at the most recent data that largest reflects reality.

One Last Thing - "Verifiability/Explainability"
Understanding how an AI works is essential to fostering trust and conviction in AI production models.

Neural Networks and Deep Learning differ from other types of Machine Learning algorithms in that they have low explainability. They can generate a prediction, but it is very difficult to understand or explain how it arrived at its prediction. This "explainability problem" is often described as a problem for all of AI, but it's primarily a problem for Neural Networks and Deep Learning. Other types of Machine Learning algorithms - for example decision trees or linear regression- have very upper explainability. The results of the five-year DARPA Explainable AI Program (XAI) are worth reading here.

So What Can Machine Learning Do?

It's taken decades but as of today, on its simplest implementations, machine learning applications can do some tasks largest and/or faster than humans. Machine Learning is most wide and widely unromantic today in processing text (through Natural Language Processing) followed by understanding images and videos (through Computer Vision) and analytics and oddity detection. For example:

Recognize and Understand Text/Natural Language Processing
AI is largest than humans on vital reading comprehension benchmarks like SuperGLUE and SQuAD and their performance on ramified linguistic tasks is scrutinizingly there. Applications: GPT-3, M6, OPT-175B, Google Translate, Gmail Autocomplete, Chatbots, Text summarization.

Write Human-like Answers to Questions and Squire in Writing Computer Code
An AI can write original text that is indistinguishable from that created by humans. Examples GPT-3, Wu Dao 2.0 or generate computer code. Example GitHub Copilot, Wordtune

Recognize and Understand Images and video streams
An AI can see and understand what it sees. It can identify and snift an object or a full-length in an image or video. It can plane identify faces. It can scan news broadcasts or read and assess text that appears in videos. It has uses in threat detection – airport security, banks, and sporting events. In medicine to interpret MRI's or to design drugs. And in retail to scan and unriddle in-store imagery to intuitively determine inventory movement. Examples of ImageNet benchmarks here and here

Turn 2D Images into 3D Rendered Scenes
AI using “NeRFs "neural radiance fields" can take 2d snapshots and render a finished 3D scene in realtime to create avatars or scenes for virtual worlds, to capture video priming participants and their environments in 3D, or to reconstruct scenes for 3D digital maps. The technology is an enabler of the metaverse, generating digital representations of real environments that creators can modify and build on. And self driving cars are using NeRF’s to render city-scale scenes spanning multiple blocks.

Detect Changes in Patterns/Recognize Anomalies
An AI can recognize patterns which don't match the behaviors expected for a particular system, out of millions of different inputs or transactions. These applications can discover vestige of an wade on financial networks, fraud detection in insurance filings or credit vellum purchases; identify fake reviews; plane tag sensor data in industrial facilities that midpoint there's a safety issue. Examples here, here and here.

Power Recommendation Engines
An AI can provide recommendations based on user behaviors used in ecommerce to provide well-judged suggestions of products to users for future purchases based on their shopping history. Examples: Netflix, TikTok, CrossingMinds and Recommendations AI

Recognize and Understand Your Voice
An AI can understand spoken language. Then it can comprehend what is stuff said and in what context. This can enable chatbots to have a conversation with people. It can record and transcribe meetings. (Some versions can plane read lips to increase accuracy.) Applications: Siri/Alexa/Google Assistant. Example here

Create Strained Images
AI can create strained images (DeepFakes) that are indistinguishable from real ones using Generative Adversarial Networks. Useful in entertainment, virtual worlds, gaming, fashion design, etc. Synthetic faces are now indistinguishable and increasingly trustworthy than photos of real people. Paper here.

Create Artist Quality Illustrations from A Written Description
AI can generate images from text descriptions, creating anthropomorphized versions of animals and objects, combining unrelated concepts in plausible ways. An example using is Dall-E

Generative Diamond of Physical Products
Engineers can input diamond goals into AI-driven generative diamond software, withal with parameters such as performance or spatial requirements, materials, manufacturing methods, and forfeit constraints. The software explores all the possible permutations of a solution, quickly generating diamond alternatives. Example here.

Sentiment Analysis
An AI leverages deep natural language processing, text analysis, and computational linguistics to proceeds insight into consumer opinion, understanding of consumer sentiment, and measuring the impact of marketing strategies. Examples: Brand24, MonkeyLearn

AI in National Security

Much like the dual-use/dual-nature of classical computers AI ripened for commercial applications can moreover be used for national security.

AI/ML and Ubiquitous Technical Surveillance
AI/ML have made most cities untenable for traditional tradecraft. Machine learning can integrate travel data (customs, airline, train, car rental, hotel, license plate readers¦,) integrate feeds from CCTV cameras for facial recognition and gait recognition, breadcrumbs from wireless devices and then combine it with DNA sampling. The result is streamlined persistent surveillance.

China's employment of AI as a tool of repression and surveillance of the Uyghurs is a reminder of a dystopian future of how totalitarian regimes will use AI-enabled ubiquitous surveillance to repress and monitor its own populace.

AI/ML on the Battlefield
AI will enable new levels of performance and autonomy for weapon systems. Autonomously collaborating assets (e.g., drone swarms, ground vehicles) that can coordinate attacks, ISR missions, & more.

Fusing and making sense of sensor data (detecting threats in optical /SAR imagery, classifying watercraft based on radar returns, searching for anomalies in radio frequency signatures, etc.) Machine learning is largest and faster than humans in finding targets subconscious in a high-clutter background. Streamlined target detection and fires from satellite/UAV.

For example, an Unmanned Aerial Vehicle (UAV) or Unmanned Ground Vehicles with on workbench AI whet computers could use deep learning to detect and locate unseen chemical, biological and explosive threats by fusing imaging sensors and chemical/biological sensors.

Other examples include:

Use AI/ML countermeasures versus adversarial, low probability of intercept/low probability of detection (LPI/LPD) radar techniques in radar and liaison systems.

Given sequences of observations of unknown radar waveforms from wrong-headed emitters without a priori knowledge, use machine learning to develop behavioral models to enable inference of radar intent and threat level, and to enable prediction of future behaviors.

For objects in space, use machine learning to predict and typify a spacecrafts possible actions, its subsequent trajectory, and what threats it can pose from withal that trajectory. Predict the outcomes of finite burn, continuous thrust, and impulsive maneuvers.

AI empowers other applications such as:

• Flight Operations Planning Visualization Aid Tool for Strike Operations Aboard Watercraft Carriers
• Automated Battle management - air and missile defense, army/navy tactical¦

AI/ML in Collection
The front end of intelligence hodgepodge platforms has created a firehose of data that have overwhelmed human analysts. "Smart" sensors coupled with inference engines that can pre-process raw intelligence and prioritize what data to transmit and store -helpful in degraded or low-bandwidth environments.

Human-Machine Teaming in Signals Intelligence
Applications with embedded intelligence have once begun to towards in commercial applications thanks to massive language models. For example – Copilot as a pair-programmer in Microsoft Visual Studio VSCode. It's not nonflexible to imagine an AI that can snift and isolate anomalies and other patterns of interest in all sorts of signal data faster and increasingly reliably than human operators.

AI-enabled natural language processing, computer vision, and audiovisual wringer can vastly reduce transmission data processing. Advances in speech-to-text transcription and language analytics now enable reading comprehension, question answering, and streamlined summarization of large quantities of text. This not only prioritizes the work of human analysts, it's a major gravity multiplier

AI can moreover be used to automate data conversion such as translations and decryptions, progressive the worthiness to derive violating insights.

Human-Machine Teaming in Tasking and Dissemination
AI-enabled systems will automate and optimize tasking and hodgepodge for platforms, sensors, and resources in near-real time in response to dynamic intelligence requirements or changes in the environment.

AI will be worldly-wise to automatically generate machine-readable versions of intelligence products and disseminate them at machine speed so that computer systems wideness the IC and the military can ingest and use them in real time without transmission intervention.

Human-Machine Teaming in Exploitation and Analytics
AI-enabled tools can plicate filtering, flagging, and triage wideness multiple data sets. They can identify connections and correlations increasingly efficiently and at a greater scale than human analysts, and can flag those findings and the most important content for human analysis.

AI can fuse data from multiple sources, types of intelligence, and nomenclature levels to produce well-judged predictive wringer in a way that is not currently possible. This can modernize indications and warnings for military operations and zippy cyber defense.

AI/ML Information warfare
Nation states have used AI systems to enhance disinformation campaigns and cyberattacks. This included using "DeepFakes" (fake videos generated by a neural network that are nearly indistinguishable from reality). They are harvesting data on Americans to build profiles of our beliefs, behavior, and biological makeup for tailored attempts to manipulate or coerce individuals.

But considering a large percentage of it is open-source AI is not limited to nation states, AI-powered cyber-attacks, deepfakes and AI software paired with commercially misogynist drones can create "poor-man's smart weapons" for use by rogue states, terrorists and criminals.

AI/ML Cyberwarfare
AI-enabled malware can learn and transmute to a system's defensive measures, by probing a target system to squint for system configuration and operational patterns and customize the wade payload to determine the most opportune time to execute the payload so to maximize the impact. Conversely, AI-enabled cyber-defensive tools can proactively locate and write network anomalies and system vulnerabilities.

Attacks Versus AI – Adversarial AI
As AI proliferates, defeating adversaries will be predicated on defeating their AI and vice versa. As Neural Networks take over sensor processing and triage tasks, a human may only be alerted if the AI deems it suspicious. Therefore, we only need to defeat the AI to evade detection, not necessarily a human.

Adversarial attacks versus AI fall into three types:

• Data misclassification– to generate false positive or negative results
• Synthetic data generation-to feed false information
• Data wringer - for AI-assisted classical wade generation

AI Wade Surfaces
Electronic Wade (EA), Electronic Protection (EP), Electronic Support (ES) all have analogues in the AI algorithmic domain. In the future, we may play the same game well-nigh the "Algorithmic Spectrum," denying our adversaries their AI capabilities while defending ours. Other can steal or poison our models or manipulate our training data.

We're Just Getting Started
Progress in AI has been growing exponentially. The next 10 years will see a massive resurgence on AI inference and training capabilities. This will require regular refreshes of the hardware- on the tweedle and deject clusters – to take advantage. This is the AI version of Moore's Law on steroids - applications that are completely infeasible today will be easy in 5 years.

What Can't AI Do?

While AI can do a lot of things largest than humans when focused on a narrow objective, there are many things it still can't do. AI works well in specific domain where you have lots of data, time/resources to train, domain expertise to set the right goals/rewards during training, but that is not unchangingly the case.

For example AI models are only as good as the fidelity and quality of the training data. Having bad labels can wreak havoc on your training results. Protecting the integrity of the training data is critical.

In addition, AI is hands fooled by out-of-domain data (things it hasn't seen before). This can happen by "overfitting" – when a model trains for too long on sample data or when the model is too complex, it can start to learn the "noise," or irrelevant information, within the dataset. When the model memorizes the noise and fits too closely to the training set, the model becomes "overfitted," and it is unable to generalize well to new data. If a model cannot generalize well to new data, then it will not be worldly-wise to perform the nomenclature or prediction tasks it was intended for. However, if you pause too early or exclude too many important features, you may encounter the opposite problem, and instead, you may "underfit" your model. Underfitting occurs when the model has not trained for unbearable time, or the input variables are not significant unbearable to determine a meaningful relationship between the input and output variables.

AI is moreover poor at estimating uncertainty /confidence (and explaining its decision-making). It can't segregate its own goals. (Executives need to pinpoint the visualization that the AI will execute. Without well-defined decisions to be made, data scientists will waste time, energy and money.) Except for simple cases an AI can't (yet) icon out rationalization and effect or why something happened. It can't think creatively or wield worldwide sense.

AI is not very good at creating a strategy (unless it can pull from previous examples and mimic them, but then fails with the unexpected.) And it lacks generalized intelligence e.g. that can generalize knowledge and translate learning wideness domains.

All of these are research topics urgently stuff worked on. Solving these will take a combination of high-performance computing, wide AI/ML semiconductors, creative machine learning implementations and decision science. Some may be solved in the next decade, at least to a level where a human can't tell the difference.

Where is AI and National Security Going Next?

In the near future AI may be worldly-wise to predict the future deportment an traducer could take and the deportment a friendly gravity could take to counter these. The 20th century model loop of Observe-Orient-Decide and Act (OODA) is retrospective; an observation cannot be made until without the event has occurred. An AI-enabled decision-making trundling might be 'sense-predict-agree-act': AI senses the environment; predicts what the traducer might do and offers what a future friendly gravity response should be; the human part of the human-machine team agrees with this assessment; and AI acts by sending machine-to-machine instructions to the small, wiry and many voluntary warfighting resources deployed en masse wideness the battlefield.

An example of this is DARPA's ACE (Air Combat Evolution) program that is developing a warfighting concept for combined stovepipe using a manned and unmanned systems. Humans will fight in tropical collaboration with voluntary weapon systems in ramified environments with tactics informed by strained intelligence.

A Once-in-a-Generation Event
Imagine it's the 1980's and you're in tuition of an intelligence agency. SIGINT and COMINT were analog and RF. You had worldwide hodgepodge systems with bespoke systems in space, air, underwater, etc. And you wake up to a world that shifts from copper to fiber. Most of your people, and equipment are going to be obsolete, and you need to learn how to capture those new bits. Scrutinizingly every merchantry processes needed to change, new organizations needed to be created, new skills were needed, and old ones were obsoleted. That's what AI/ML is going to do to you and your agency.

The primary obstacle to innovation in national security is not technology, it is culture. The DoD and IC must overcome a host of institutional, bureaucratic, and policy challenges to raising and integrating these new technologies. Many parts of our culture are resistant to change, reliant on traditional tradecraft and ways of collection, and uneager to risk-taking, (particularly acquiring and raising new technologies and integrating outside information sources.)

History tells us that late adopters fall by the wayside as increasingly wiry and opportunistic governments master new technologies.

Carpe Diem.

Want increasingly Detail?

Read on if you want to know well-nigh Machine Learning chips, see a sample Machine Learning Pipeline and learn well-nigh the four types of Machine Learning.

Details of a machine learning pipeline

This is a sample of the workflow (a pipeline) data scientists use to develop, deploy and maintain a machine learning model (see the detailed unravelment here.)

(download a PDF of this vendible here)

Sources:

• Understanding AI Technology: Greg Allen, Chief of Strategy and Communications Joint Strained Intelligence Part-way (JAIC), Department of Defense https://www.ai.mil/docs/Understanding AI Technology.pdf
• ·AI, Machine Learning, Deep Learning Explained Simply: Jun Wu https://towardsdatascience.com/ai-machine-learning-deep-learning-explained-simply-7b553da5b960
• The Democratization of Strained Intelligence and Deep Learning: Databrickshttps://databricks.com/discover/pages/the-democratization-of-artificial-intelligence-and-deep-learning
• Final Report: National Security Report on Strained Intelligencehttps://www.nscai.gov/wp-content/uploads/2021/03/Full-Report-Digital-1.pdf
• A Beginners Guide to Neural Nets and Deep Learning: Pathmind https://wiki.pathmind.com/neural-network
• IBM Deject Learning: Overfiitting https://www.ibm.com/cloud/learn/overfitting