Home » Markets » AI (page 7)

AI

The support or use of AI (artificial intelligence) in electronics, including ML (machine learning), whether in software (supervised, unsupervised or reinforcement learning tools) or hardware (accelerators, GPUs, etc).

Perceptra raises €1.2m

IMG_1533-300x200.webp

Perceptra, the MIT spin-off pioneering photonic chip-based AI-enabled Raman sensors for chemical monitoring, has secured a €1.2 million investment from PhotonDelta, the Dutch photonic chip industry accelerator. The funding will accelerate the development and commercialisation of its next-generation of photonic chip-based Raman sensors. “PhotonDelta’s funding marks a milestone moment in turning our research into a scalable venture,” says CEO Amir ...

The application areas where RFID and AI converge

RFID-300x200.webp

AI is being paired with RFID technology across many industries and applications, observes Liam Critchley. Many industries need to track people or objects efficiently, autonomously and remotely, and that’s where radio frequency identification (RFID) comes in. It provides real-time tracking, which requires the accurate collection and sharing of data. AI is now being paired with RFID technology to help access ...

Design for ethics: hardware’s role in AI alignment

Ethics-1-300x200.webp

Silicon’s potential for mitigating risks in artificial intelligence is underrated, argues Paul Dempsey. The concept of ‘known unknowns and unknown unknowns’ was coined by researchers at Nasa and the Rand Corporation in the 1950s and ’60s to assess risk. The same approach is being applied to the design of AI systems. Because of concerns over how generative AI (GenAI) performs ...

HBM4 set back a quarter for re-design

IMG_1561-300x200.webp

HBM4 mass production has been put back to the end of Q1 at the earliest, TrendForce, after Nvidia increased the spec for the pin speed for its Rubin platform to over 11Gbps requiring a redesign by Hynix, Samsung and Micron. All three companies are said to have submitted re-designed samples. Samsung is said to be utilising a more advanced process ...

Datacentre liquid cooling market to reach $7bn by 2029

IMG_1559-300x200.webp

The Data Centre Liquid Cooling market is expected to grow rapidly to about $7 billion in manufacturer revenue by 2029, says Dell’Oro. The expansion reflects liquid cooling’s transition from a niche technology to a foundational requirement for deploying AI infrastructure, as rising accelerator TDP and rack densities push air cooling beyond practical limits. “Liquid cooling has crossed a critical threshold,” ...

Cadence and IP partners provide pre-validated packaged chiplets

Cadence chiplet ecosystem

At CES 2026, Cadence announced an ecosystem which delivers pre-validated (spec-to-packaged parts) for physical AI, data centre and HPC (high performance computing). In Las Vegas, Cadence announced that Arm, Arteris, eMemory, M31 Technology, Silicon Creations and Trilinear Technologies are IP partners for the Physical AI chiplet platform and chiplet framework. The ecosystem also includes silicon analytics partner proteanTecs. A silicon ...

JEDEC preps SPHBM4 for HBM4-level rates with reduced pin count

AI-datacentre-300x200.webp

The JEDEC Solid State Technology Association says it is nearing completion of a new standard for Standard Package High Bandwidth Memory (SPHBM4). The development addresses the increasing memory requirements of AI data centres. As planned, SPHBM operates at the same aggregate data throughput as HBM4. But it uses fewer pins by operating at a higher frequency. SPHBM4 SPHBM4 devices are ...

Nvidia launches Rubin

IMG_1546-300x200.webp

Nvidia has launched the Rubin platform comprising six new chips designed to deliver one AI supercomputer. The platform uses extreme codesign across the six chips – the NVIDIA Vera CPU, NVIDIA Rubin GPU, NVIDIA NVLink 6 Switch, NVIDIA ConnectX-9 SuperNIC, NVIDIA BlueField-4 DPU and NVIDIA Spectrum-6 Ethernet Switch – to slash training time and inference token costs. “Rubin arrives at ...

Furiosa AI to start mass production this month

IMG_1528-300x200.webp

Furiosa AI, the Korean AI chip startup, plans to start mass production of its first IC this month. Furiosa AI was founded in 2017 by June Paik (pictured) formerly of AMD and Samsung and named after the Imperator Furiosa – a character in the 2015 film Mad Max: Fury Road. Furiosa’s accelerators use an architecture it calls Tensor Contraction Processor ...

Chinese AI IC developer Biren IPOs in Hong Kong

IMG_1534-300x200.webp

Another Chinese AI chip developer, Biren Technology, has had a stellar IPO. Biren shares rose closed 75% over the offer price on the Hong Kong stock exchange on Boxing Day, reports the Nikkei. Biren raised $716.85 million and achieved a market cap of $10 billion. Since Jan. 1, 2022, Biren has accumulated losses of $486 million. At of  H1 2025, ...