AI has the potential to change how connectors are made and who makes them.

For the future of the electronics industry, especially the connector industry faced with what kind of difficulties, we should be open to all rivers, seek truth from facts. The most important change right now is artificial intelligence (AI), not new processors, fast interconnects, good sensors or new memory chips, which are secondary. The real question is how the connectors industry will cope with the flood of revolutionary technology, legitimate patents, competitiveness, and standard issues that will inevitably arise from the flood of artificial intelligence. Change is accelerating and the connector industry needs to face up to it.

The connector industry follows four basic principles:

1. Connector manufacturers should understand connectors better than customers.

2. Only connector manufacturers have the skills and equipment for mass production.

3. Use your patent portfolio to protect your business model.

4. Connector manufacturers should ensure the reliability of their products in the use process to avoid the resulting application failure.

Now, let's take a closer look at these principles, while acknowledging the real threat posed by AI. Russian President Vladimir Putin once said in a school speech: "Artificial intelligence is not only the future of Russia, but the future of all mankind. Whoever becomes the leader in this field will be the ruler of the world." Of course, he was talking mainly about intelligent weapons and battlefield robots. But does the same apply to connectors? Let's explore how this possibility affects the foundations of the industry.

The Power of Data

Connector manufacturers spend large sums of money developing simulation models for their high-speed copper connectors to determine performance parameters and eye diagrams associated with certain chips (e.g., PCI Express, Ethernet) and transmission lines (e.g., cable, board, PCB lines). They share this data with the standards committee. The basic premise of the industry is that connector manufacturers understand connectors more deeply and thoroughly than customers. But AI has the potential to change that premise.

AI is not about having the best computing power. It's about having a huge amount of data. And top system vendors use this huge amount of data to predict connector performance and reliability. System vendors build thousands or millions of systems, and through the experience curve, they have a good idea of cost, performance, and reliability. Through this data, system developers of artificial intelligence computer-aided design (CAD) systems know better how to design connectors. They just don't have the capacity to produce it.

How does a connector manufacturer get data from a system vendor? They can pay for AI-CAD tools and servers and offer them as a service to their customers, who don't have to buy them themselves. That way, they can get the data for free and get the cost of using the tool from the customer. This process is known as software services or SaaS. Connector vendors' tools also collect a lot of data about how customers use their connectors, another sign of the power of the Internet. System vendors know that their data is of great value, and leveraging the connector design elements created will have a competitive advantage in the marketplace.

Personally, I don't think the SaaS model works for connector manufacturers unless they combine and separate their own manufacturing expertise data with their customers' operational and reliability data.

As AI-CAD matures, the tendency for customers to share databases with connector vendors may diminish. Where we're likely to go: At some point databases may be more valuable than patents, and the connectors industry's patent-based business model may be threatened. But if you think AI-CAD can surpass the patent system, it is not possible.

System vendors now have better data than connector manufacturers on how to design good connectors. How do they use AI-CAD data? When a powerful 3D printer is available, they will connect the system to the 3D printer and make their own connectors. AI-CAD and 3D printers are really two of the best tools for connector manufacturers.

Missing Standards

Let me be clear: I'm not a lawyer. The information provided herein is not intended to convey or constitute legal advice or advice. I'm just asking questions and ideas. If you have questions or concerns, you should contact the relevant attorney.

Much has been written about how AI will affect the economy, society, health care, the financial industry, and online employment. But I couldn't find a single article on how AI affected standards development. I found no documentation or ideas from ANSI or any other major standards development organization (SDO). This is both disappointing and unsettling. So, we're going to highlight that here.

Go to any standards development conference and you're likely to run into a number of engineers from connector companies. Connector companies operate on open hardware standard hardware for the telecommunications, military, industrial, transportation, and medical markets. They use intellectual property, such as patents, to protect their business.

Assuming that a company has a good AI-CAD system and is involved in the standard development process, they can introduce this new standard information into the system and prepare the basic elements of the new standard in advance. The AI-CAD system can then make applications and patent applications for each of the basic elements. This is nothing new. For decades, attendees to standards conferences have used the sessions to establish design features, parameters, etc., and file for patent protection. AI-CAD will speed up this process, potentially allowing companies involved in the development of standards to scramble and determine the standards process.

The United States has been the No. 1 country in patent filing since September 2011. Every time a standard is developed, several US connector suppliers will declare a patent (or patent application) at a meeting. There has been a lot of discussion about the first publication rather than first filing system in the new patent law. This includes grace periods and subsequent disclosures that are not considered "prior techniques" and so on. This process can take months or years. Depends on the complexity of the documents and the legal process. AI-CAD will exacerbate SDO and participate in the discussion of the company's current patent standards.

SDO might add an AI system to its cloud system and charge per view. They can upload the scope and parameters of new standards, and these algorithms can also determine the probabilities and fundamentals of system branching. In other words, AI systems can actually write standards. It can also identify all relevant standard basic patents (SEPs) and circumvented patent infringement by optimizing the branch system, except for patent applications, which are not publicly disclosed for 18 months. Such a process may anger participating company members and may cause some legal problems.

SDO members can also link their AI-CAD systems together in a collective agreement and create their own standards. They can bypass SDOs and guilds entirely. Linking SDO's AI machines to its members' AI machines will produce some interesting legal, technical, social, and economic things that we have never seen before. Can the Department of Justice legally authorize AI to allow any AI machine to connect to the SDO network? We have shown that AI machines can think faster than engineers, and even they can think faster than lawyers. Lawyers are extremely busy in today's rule of law society, and their mission in life is, in part, to be at the forefront of innovation, technological development, and experimentation. The problem of integrating AI-CAD and law is like the problem of mixing oil and water.

Slaughter - bots and responsibility

Slaughter- Bots: A Bionic AI Drone for Killing Criminals

Watching the Slaughter- Bots video shows that the robot is entirely powered by sophisticated AI. Future military forces and law enforcement agencies can take advantage of these new tools. But the assumption that these devices kill innocent bystanders in the course of their missions could lead to civil and criminal prosecutions. But who will be prosecuted or charged? The team that sent the robots? Robot makers? The person or company that wrote the algorithm? Providing the owner of the robot database? A company that provides robot components? Or the company that passes data from sensors to processors? The answers to these questions are difficult to determine.

At a lower application level, if an AI system in SDO does a patent search, determines the best path based on the underlying standard subsystems, but still misses some patents or applications. Is SDO liable to litigants who rely on search-based patents? Just as there is a lack of papers on the development of AI standards, there is a lack of research papers on potential AI responsibilities and complex problems in the legal community.

The problem of talent

Designing connectors is a problem of limited scope. The three-dimensional design of connectors will tie us down; The conductive and insulating materials used for enclosures and contact points will bind us; The laws of physics and transmission lines that use copper and fiber optics also bind us. Talent to break through these constraints is scarce, and the level of income of such talent is very high. This is why many programmers and engineers work hard on these issues, especially in the AI-CAD field.

For a new product that takes engineers months to design, AI-CAD systems with massive amounts of data can be completed in minutes or hours. The AI system can then verify the design in a simulation within minutes. There is no need to build multiple models for expensive testing and design revisions. So, are we connectors, semiconductors, circuit boards and other traditional engineers out of work in the future? I think the answer is yes, just not to what extent.

conclusion

Why would Amazon buy Whole Foods to get into the food business and not connectors to get into the connectors business? In the United States alone, the food industry generates about $16 billion in sales every day. The connector industry generates about $164 million in sales worldwide every day. For Amazon, the connector business is limited. But for small CAD product companies, it is worth a try. Amazon now has more than 2m servers installed in its data centres. Their staff have designed their own servers and built them by manufacturers in Asia. If anyone has a huge database of connector costs, performance, and reliability, it's Amazon, Google, or Facebook. Can they run this data through an AI system and design better connectors and have them made in China? To some extent, they probably already have.

Ray Kurzweil, Stephen Hawking, Elon Musk and other futurists have commented that artificial intelligence will take humanity to the surprising point where robots will become smarter than humans. They predict the event will take place around 2045. But they may be wrong about the timing.

This will happen when AI-CAD is faced with legal and regulatory confusion, standard-setting, patenting systems, and the transfer of data from manufacturers to customers.

A cursory look at a few elements of this theme reveals that AI-CAD may have had a more significant impact on the industry than this article suggests. If I had a good database and predictive AI algorithm, I could show you a tree of what's likely to happen, showing what each branching system is capable of. Examine the impact of primary, secondary, and other factors to determine where the impact is on the industry.

But first, we have to admit that even as we ruminate, the AI may already have an answer. While we spend hours poring over spreadsheet data to find answers, the AI can manipulate hundreds or thousands of variables and perform millions of calculations in seconds to arrive at a result. Although we are only speculating, the AI may already know where the future conclusion lies.