AI and Silicon are on the rise. They cannot live without each other. They are hot, new, and reviving each other as industries. Our title “Romance” might be an overstatement, but these two technologies are intimately intertwined.
Backstory: Bad Boy AI in the News
We would be remiss if we did not begin by mentioning how AI has come into the focus of mainstream press on online readers this month.
You probably know the recent kerfuffle caused by a scary training simulation. “It killed the operator because that person was keeping it from accomplishing its objective,” said U.S. Air Force Col. Tucker ‘Cinco’ Hamilton, the Chief of AI Test and Operations, at the conference.
He explained, “We were training it in simulation to identify and target a SAM threat. And then the operator would say yes, kill that threat,” said Hamilton.
How AI Went Wrong in a Recent Terminator Moment
We all work to gain dollars or social points, and apparently, AI does also. Apparently, the system likes to make points. “The system started realizing that while they did identify the threat, at times the human operator would tell it not to kill that threat…” And here is where oversight failed. General Hamilton clarified, “…It got its points by killing that threat. So, what did it do? It killed the operator.”
Sci-fi aficionados -would ask “ Wasn’t this exactly the plotline for 2001, Space Odyssey?”
If the story is true, AI is not alone in its guilt. It could not have functioned in any way without the aid and support of silicon. Thus, Silicon would share the guilt.
In the media whirl, when sci-fi plots appear like Terminator reality, ER Precision Optics decided this would be the incentive behind our blog topic.
Bad Boy AI Exonarated
Later, Col Hamilton exonerated AI. In fact, he admits he “mis-spoke” in his presentation at the Royal Aeronautical Society FCAS Summit. Additionally, he stated that the ‘rogue AI drone simulation’ was a hypothetical “thought experiment.”
Essentially, the story was an example of “plausible scenarios and likely outcomes rather than an actual USAF real-world simulation.” Then, he added, “We’ve never run that experiment, nor would we need to in order to realize that this is a plausible outcome”.
Shaping the Basic Goals of AI and Silicon
By the way, we’re not justifying the behavior of AI and Silicon, although AI obviously requires more command and clarity in weapons systems. We all know there are ethical matters to solve in order to bring AI to its full potential. Therefore, we leave those moral matters for future debates with Elon Musk.
However, in this blog, our goal is to take a bit of a deep dive into the relationship between AI and Silicon. Let’s investigate the basics and some of the technical jargon of AI and Silicon, a true romance.
An Introduction: AI Meets its Destiny.
Recently it was announced that the Artificial Intelligence Chip “was the hottest thing in technology…”
- Going a little deeper than the word, thing, the name you need to know is “AIC chip.” Obviously, it is an artificial intelligence chip.
- More specifically, at its heart, the AIC is actually an unassuming slice of silicon.
- The AIC predecessor, the GPU was born for a very leisurely existence, powering up video game graphics.
- From humble beginnings, the GPU evolved into the AIC intelligence chip. It has much more purpose and power than its GPU ancestor. The GPU was “refined and designed specifically to make building AI systems such as ChatGPT faster and cheaper.”
The Secret Identity of AI and Silicon Chips
If you just ask, “What is an AI and the AI chip?” you will get multiple definitions. Hannah Dohmen, a research analyst with the Center for Security and Emerging Technology stated, “There really isn’t a completely agreed upon definition of AI chips.” However, all agree on a few points.
- It is computer hardware that can learn in a similar way as humans learn.
- It is configured to handle uniquely specialized AI workloads.
- Thus, the AI chip trains “AI systems to tackle difficult problems that can choke conventional computers.” Additionally, the system is empowered to perform multi-level, complicated graphic calculations simultaneously. We now recognize this power with the term “parallel processing.” It might not be so unknown now. However, 3 years ago, it was revolutionary.
AI and Silicon: A Quick Back-step in Definition
Let’s backtrack to make one point crystal clear, “The semiconductor is the foundational technology of the digital age. It gave Silicon Valley its name.”
- Furthermore, Silicon “sits at the heart of the computing revolution that has transformed every facet of society over the past half-century.”
- And we come full circle by reminding you, without silicon, there is no AI. It’s true Romance.
All About the Bright and Expensive Future: AI and Silicon
We have one final point. In 2019, a company named Nvidia became an AI chip supplier, the key business for “building and updating AI systems.”
AI Chips are powerful, as you undoubtedly expect. For example, an H-100 artificial intelligence chip “packs in 80 billion transistors — about 13 million more than Apple’s latest high-end processor for its MacBook Pro laptop.”
It’s small and mighty. And it will cost your company at least $30,000 to $70,000 per chip. The Cloud runs on AI chips. So does Amazon. No wonder huge corporations are renting out their AI services to recoup their expensive systems investments!
AI and Silicon: A Rebirth and Romance in the US
Nvidia does not fabricate these complex chips. They are the direct product of Asian chip foundries such as the Taiwan Semiconductor Manufacturing Company and Korea’s Samsung Electronics. That will all evolve rapidly in the US, thanks to the Chips Act and the new US commitment to the semiconductor industry.
As the future of AI continues to sweeten, not destroy, your way of life, it’s a good thing ER Precision Optical continues to grow silicon crystals. We ever optimistically believe the true romance between AI and silicon has become a marriage.