IBM advances Quantum Computing ambitions with generative AI power for software development

Are you ready to bring more awareness to your brand? Consider becoming a sponsor for The AI Impact Tour. Learn more about the opportunities here.

Quantum computing is taking another ‘quantum’ leap forward today with new processor, system and software updates announced at the IBM Quantum 2023 Summit.

The star of the announcements is the IBM Quantum Heron processor, seen below in an image provided to VentureBeat by IBM.

IBM Quantum Heron chip. Credit: IBM

With 133 qubits, IBM claims that Heron is its highest-performing quantum processor to date, offering significant improvements in qubit quality and error rates compared to previous IBM quantum chips. Previously the IBM Quantum Eagle was the most powerful active system with a demonstrated power of 127 qubits. The qubit is a defining metric of quantum computing and a bigger number translates to more performance.

Going a step further, IBM announced that it is now developing an even more powerful chip known as Condor that will have 1,121 superconducting qubits on a single chip

VB Event

The AI Impact Tour

Connect with the enterprise AI community at VentureBeat’s AI Impact Tour coming to a city near you!


Learn More

Along with the Quantum Heron, IBM announced that its Quantum System Two architecture, which is powered by three IBM Quantum Heron processors has begun operations. The first IBM Quantum System Two machine is installed at IBM’s quantum computing hub in Yorktown Heights, New York.

To help researchers and enterprises alike benefit from the power of Quantum systems, IBM also detailed innovations to its Qiskit quantum programming software that will benefit from the power of generative AI.

“The full power of using quantum computing will be powered by generative AI to simplify the developer experience,” Jay Gambetta, VP at IBM Quantum explained during a briefing with press and analysts.

How Qiskit is using generative AI to enable Quantum computing programming

Programming a quantum system requires different software than a classical computing system. That’s where the IBM Qiskit programming framework comes into play. 

IBM Quantum System. Credit: IBM

At the IBM Quantum Summit 2023, the company detailed features in Qiskit 1.0, which is set to become available in early 2024. Gambetta explained during the briefing that part of the 1.0 release is the concept of Qiskit patterns. He noted that it’s not practical for developers to have quantum circuit knowledge to be able to do work.

“We’ve come up with a simple strategy or framework to develop an algorithm, it consists of a way of mapping a problem to quantum circuits and operators,” Gambetta said.

Qiskit patterns will also optimize the problems for quantum execution and run them on a system like the IBM Quantum System Two, providing processing results with a simple output. To simplify the process even further, IBM will also be taking generative AI tools from the company’s Watsonx portfolio to enable simple language commands to generate a quantum circuit. Watsonx provides a foundation and large language models (LLM) to help solve different use cases, including application development. IBM is already using Watsonx to help modernize the COBOL programming language on mainframes for the modern era.

In response to a question from VentureBeat, Gambetta explained that IBM is using the same Watsonx granite foundation models it has been building out this year for other use cases, for the Qiskit quantum use case.

“This is one of the exciting things about what the team has done with Watsonx, it’s actually the same model, now you can fine-tune it based on Qiskit,” Gambetta said. 

AI is also being used by IBM to improve the optimization of quantum circuits. Gambetta added that moving forward he expects to see AI coming to quantum to improve how quantum works.

Quantum isn’t just for research, it’s for enterprises too

Quantum computing is often thought of as being just about very advanced research use cases. According to Gambetta, quantum computing also has a place in enterprises as well.

“We already have many enterprises working with us today,” he said.  “In my opinion, we’re in the most exciting time of technology where research and commercialization is closer than ever.”

Gambetta said that today IBM has more than 160 client industry clients that have been either working with IBM or its partners on enterprise experiments. While there is lots of activity, it’s still very much a work in progress.

“I’m not gonna sit here and say they got a return on investment yet, but they’re actually starting to transition from just being quantum ready, to actually doing use case prototypes,” Gambetta said.

He noted that one of the demonstrations that is being shown at the IBM Quantum Summit this year is Hyundai running a very large optimization problem. The challenge that IBM is working on is discovering the algorithms that are needed to advance different use cases as well as improving the performance of quantum hardware and software.

“We’re committed to making these discoveries possible, but like it takes work,” Gambetta said.

VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.


Leave a Comment