Designed By Humans, Perfected By A.I.

China has stuck its fingers into automating chip design using artificial intelligence — taking it a step further. It may not be so if one looks deeper; it is still designed by humans, only “perfected” by A.I.. 

DIGITimes reported scientists from the Institute of Computing Technology under Chinese Academy of Sciences have recently published a research paper showcasing capability to use AI in automating chip design. Titled “Pushing the Limits of Machine Design: Automated CPU Design with AI”, the research claims that a 32-bit RISC-V CPU design was automatically generated within 5 hours, and that the CPU, with a clock rate up to 300MHz, already entered tape-out in December 2021 on 65nm process.

In addition, the research claimed that the CPU was tested successfully on the Linux operating system and performed “comparably against the human-designed Intel 80486SX CPU.”

This came shortly after Nvidia’s CEO Jensen Huang declared his take about the power of A.I. — qualifying anyone as a programmer. No longer will programmers need to write lines of code, only for it to display the dreaded “fail to compile” because of a missing semicolon. 

How will this be done? By generative artificial intelligence, Huang said during his keynote speech at the Computex forum in Taiwan on 29 May 2023.

Genesis Remains From Engineers

Though modern commercial electronic design automation (EDA) tools such as logic synthesis or high-level synthesis tools are available to accelerate the CPU design, all these tools require hand-crafted formal program code as the input, the research paper explained. In other words, engineers must use formal programming languages to implement the circuit logic of a CPU based on design specification, before various EDA tools can be used to facilitate functional validation and performance/power optimization of the circuit logic.

The highly complex process typically iterates for months or years, “where the key bottleneck is the manual implementation of the input circuit logic in terms of formal program code,” indicated the paper.

What the research group aimed for was automating the CPU design without human programming. It was done by using partial input output examples to replace human-written programs as the inputs, as such examples were directly accessible from a large number of legacy test cases.

Therefore, according to the research, “The problem of automated CPU design can be formulated as generating the circuit logic in the form of a Boolean function satisfying the input-output specification.” Using the approach, the report claimed to successfully generate “a large-scale Boolean function with almost 100% validation accuracy (e.g., > 99.99999999999% as Intel) from only external input-output examples”.

Notably, the paper claimed that the adopted approach could discover not only the general von Neumann architecture, but also fine-grain architecture optimisation from scratch. Despite the claimed success, this only marks the first step.

If the design complexity was increased, with a more complex CPU and some other IP blocks, tougher performance targets have to be met to evaluate how good the AI flow is, according to Woz Ahmed, Managing Director of Chilli Ventures and former Chief Strategy Officer of Imagination Technologies.

“Setting aggressive design constraints with a new process and integrating complex IP is the real test, to see if automation will match experienced engineers in power, performance and area (PPA), testability and manufacturability,” Ahmed observed. “Given the US chokehold on EDA tools, China’s best hope in indigenous EDA is to leverage its broad AI capabilities,” Ahmed further noted, “they need to be getting hold of lots of relevant training data.”

The Real Battle

This research appears to be a direct attempt to discredit Nvidia CEO’s take on technology’s ability match human intelligence. This is in spite of Nvidia’s recent unveiling of their super computers.

Such is debatable given the recent prominent blunders caused by people using A.I. chatbot for programming. The value add of chatbots remains unquestionable as at today.

Despite the headline of its ban in many nations, the advanced countries recognise the advantages ChatGPT has to offer albeit the plethora of data security issues. Many organisations have resorted to develop their inhouse A.I. chatbot to prevent confidential data leak.

In fact, it may take competition a notch higher. Hence, if coding takes centrestage, the real battle is between the engineers’ coding prowess, instead of the program.

While the technology itself is impressive, it is the skill and expertise of the engineers that ultimately determine the success and capabilities of the chatbot. These engineers are tasked with training, fine-tuning, and optimising the model, ensuring that it can generate accurate and coherent responses to a wide range of queries.

They constantly strive to improve the system, addressing its limitations, enhancing its understanding, and refining its ability to engage in meaningful conversations. It is their collective knowledge, experience, and ingenuity that fuels the competition, as they push the boundaries of what ChatGPT can achieve, shaping the future of conversational AI.

Related Stories:

A.I. Powered Chatbot Explodes Globally Despite Massive Controversies
Companies Struggle To Protect Corporate Secrets From ChatGPT
Samsung Employees Accidentally Leaked Company Secrets Via ChatGPT
Samsung May Have A Semiconductor Factory In Vietnam
Samsung Sells Record US$1.7 Billion Of Phones Over India Holiday
Samsung’s Smartphone Shipments Drop By 8 Percent In Global Markets In Q3 2022
Samsung Envisions Hyper-Growth in Memory and Logic Semiconductors Through Intensified Industry Collaborations
Samsung To Produce Semiconductor Parts In Vietnam In 2023
Fine-Tuning Production With Behringer Saws Inc

WANT MORE INSIDER NEWS? SUBSCRIBE TO OUR DIGITAL MAGAZINE NOW!

CONNECT WITH US:  LinkedIn, Facebook, Twitter

Letter to the Editor
Do you have an opinion about this story? Do you have some thoughts you’d like to share with our readers? APMEN News would love to hear from you!

Email your letter to the Editorial Team at [email protected]