By Junko Yoshida

What’s at stake:
Technologies to build 2D and 2.5D multi-die chips have existed for almost a decade. Yet, demand for chiplets languished until the dawn of the generative AI era. Nvidia’s AI processors have swept the market, triggering a change so abrupt that foundries and Outsourced Semiconductor Assembly and Test (OSAT) companies are left flat-footed. At stake now is how long aspiring chiplet vendors can wait, and whether going to China for chiplet production is an option.

As global chiplet demand soars, the shortage in production capacity has focused sharply for developers of AI processors, high-performance computing chips and automotive OEMs/Tier Ones looking for scalable, automotive semiconductor designs.

By Junko Yoshida

What’s at stake:
Robotaxis have gone through cycles. Some are already out of business. If you believe AV trucks have better commercial prospects, think again. The roster is limited and the playing field hasn’t been chalked. Compared to robotaxis, AV trucks are still in their infancy.

Last Friday, California Governor Gavin Newsom vetoed a state law requiring a trained safety driver to occupy autonomous trucks weighing over 10,000 pounds.

The governor’s action was spun differently across the political spectrum. Most vocal were AV industry supporters and lobbyists, who claimed a major victory over the labor-backed Assembly Bill 316. 

By Junko Yoshida

What’s at stake:
We have heard endlessly how robotaxis are safer than human-driven vehicles. In Vienna, at the Autonomous conference, held under the auspices of TTTech Auto, we came not necessarily to debunk that myth but to pick the brains of automakers and regulators about the thorny issues of liability.  

It’s a sin to squander an opportunity to take the stage and ask direct questions to people in charge. Waffling and evasion are typical results. Nonetheless, even those non-answers live on in the public record.

By Junko Yoshida

What’s at stake:
One boneheaded move by Nvidia to buy Arm, exacerbated by the geopolitical tension between China and the United States, forced the formation of a Europe-based RISC-V venture. 

Europe is angling to play a crucial role in the global semiconductor world,  this time with RISC-V. The formation of a new RISC-V venture – announced in August – signals the necessity of this European initiative.

By Peter Clarke

What’s at stake:
The semiconductor and broader financial markets want to draw a line under a couple of years of lackluster performance, caused by political turmoil and economic recession. If successful, an IPO by Arm – one of very few in recent quarters – could signal that confidence is ready to return, helping unleash pent-up demand for deals and technology sales optimism. Wall Street and the markets could be “off to the races.”

Arm Ltd. is set to go public but the transaction is dogged by controversy – just like almost everything about the semiconductor IP vendor in recent years. A successful debut will be cheered by Wall Street, Arm’s customers and other chip enterprises. If the IPO has to be pulled or priced low to gain support, however, it could signal that caution will prevail in the markets generally. Worse, still, SoftBank Group and Arm may be perceived as yesterday’s heroes.

By George Leopold

What’s at stake:
Cloud and telecommunications networks are groaning under the weight of ever larger data sets. Expanding optical networks are evolving to provide greater bandwidth and faster interconnects needed to move data from one facility to another. Optical chip makers and their networking equipment partners are leveraging emerging interoperability standards to help service providers super-charge their networks while avoiding vendor lock-in.

Cloud data center operators and stodgier but essential telecommunications carriers are beginning to see the light as they evolve their optical network architectures to address ever-growing bandwidth demands while extending the reach of those networks.

By Junko Yoshida

What’s at stake:
Tech companies are promoting myriad claims that they have solutions to process transformer algorithms better than others. The benchmarking of transformer engines is not yet available.

The burgeoning trend toward generative AI has flipped the whole AI world on its head, or so it seems.

Large Language Models (LLMs), as seen in ChatGPT, are mostly limited to language modeling and text generation. But transformers – an overarching deep-learning architecture that underlines LLMs and other generative AI applications – offers a model useful in data streams ranging from text, speech and image to 3D and video, or any sensory data.