DuckMan (@DuckStockMan) header
DuckMan (@DuckStockMan) profile picture

BrainChip investor

Wall Street
Joined April 2026
Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)Photo by DuckMan (@DuckStockMan)
I'm going to call this DragonFly. Four BrainChip Akida neuromorphic chips on one PCIe switch, forming a neuromorphic RAID. Most don't know that the AKD1000 has an ARM Cortex-M4 chip. I finally had some time to take advantage of this part of the chip's architecture, something I didn't do with the demos. I made the fourth chip's onboard M4 a controller for the other three, over the switch, with the host CPUs out of the loop. Lose a chip's model? The pod rebuilds it bit-exact from parity. On silicon. RAID, but for neuromorphic chips instead of disks, another first for the neuromorphic world. Three chips hold model programs. The fourth holds XOR parity and becomes a hot spare, its ARM M4 pulls the others' data over the switch, builds parity, scrubs for corruption, and reconstructs a lost chip from the survivors. The chip's own onboard microcontroller drive peer chips across the PCIe switch. Proven, bit-exact: write to a peer, read from a peer, a mailbox + doorbell protocol between the chips' M4s, all with the server CPU sitting it out, automatic failover and redundancy. So, the RAID automatically protects what the chip LEARNED. Akida learns on-chip, in the field, with no labels. I ran a real learning step on one chip and the pod journaled that exact learned change and folded it into parity. Lose the chip, keep the lesson. So a neuromorphic chip that learns something in the field can have that learning reconstructed onto a replacement from its neighbors. The model's redundancy lives on the card, not somewhere else. Where and how this would fit in the real world: * On-site learning where you can't afford to lose it: drones, remote sensors * Disconnected or contested environments: no phoning home to re-download a model * Space and radiation: bit-flips and chip loss are expected; on-card scrub + rebuild is the answer, and neuromorphic is already low-power * Industrial and medical appliances: model integrity survives a chip fault with no service call. * Large multi-chip neuromorphic fabrics as they scale: host-free resilience between chips becomes a requirement, not a luxury.
2
4
1
25
675
DuckMan retweeted
🧵🚨➡️ Quick 2026 update on the 2021 West Point CLAWS piece. Neuromorphic chips (Intel Loihi / BrainChip Akida) now appear in defense research for low-power drone detection, UAV autonomy, electronic warfare, and sensor processing. Raytheon has run student competitions using them for fully autonomous landing tasks. Still mostly edge AI and efficiency gains, not fielded systems that make independent lethal targeting calls as the original article sketched. The core “brain-like discretion under LOAC” idea remains research-stage.
🚨➡️ A 2021 West Point piece looks at “Cognitive Lethal Autonomous Weapons” (CLAWS). The idea is to use brain-mimicking AI (neuromorphic computing) so future autonomous weapons could adapt on the fly and make better targeting calls in messy urban fights. They still can’t feel emotion or honor, and the tech was still in development. lieber.westpoint.edu/cogniti…
1
2
3
112
This is a PCIe switch, it now holds 4 AKD1000's in one PCIe slot on an Intel i3-3220, 3.30 GHz, Ivy Bridge via an old HP-branded Foxconn 2ABF motherboard. The switch can actually do PCIe Gen3 but here is limited only by the CPU to 2.0. Basically, this means we can use multiple Akida cards in virtually any PCIe slot. Where we can put four, we can put many more.
5
30
359
Replying to @vip92226041
Jensen should buy Brainchip and develop neuromorphic chips to take out Nvidia/GPU.
1
1
1
63
9/10 Can something replace today’s chips? Photonic computing, analog computing and neuromorphic chips are interesting possibilities. They could make certain computations faster or more energy-efficient.
1
1
3
35
DuckMan retweeted
To the @BrainChip_inc community! What’s your take on the latest developments surrounding renewBRN and yesterday’s board resignations? Are you backing BRAINCHIP and its current direction, or do you support renewBRN and its proposed changes? Let’s hear your #thoughts. #Opinion
2
1
6
251
This is now documented as a technical paper, you can find it here, citable via the DOI on Zenodo: kevindjohnson.org/Forty-Test…
When I say the use cases for Akida are endless, I mean it. Run Akida anywhere. Run Akida everywhere.
10
34
579
When I say the use cases for Akida are endless, I mean it. Run Akida anywhere. Run Akida everywhere.
6
16
4
72
2,605
DuckMan retweeted
Interesting turn of events in the RenewBRN saga at BrainChip! Let’s see who they bring to the table… @computeontology any Idea? I…I don’t know… asx.api.cmfyapp.com/asx-rese…
2
1
17
594
Alf and Gilles are back from #electronicafair 2026 in Munich, where Physical AI was the headline. Their top takeaway: every joint and fingertip needs its own sensing, local, always on, running at milliwatt power. Event-based vision is ready for real products too.
3
14
77
1,308
Somebody sent me an AI-generated comparison that sets my neuromorphic work against Sparxity (formerly SpiNNcloud). It praises the engineering, but calls my work a “hack,” sends Akida back to the edge, and gives the data center to Sparxity. The argument rests on one presupposition: a chip can only do what a marketing brochure might say. That’s taxonomy passed off as physics. A chip’s capability belongs to the silicon, not the marketing approach. Take the main objection, that the AKD1500 has only 1 MB of memory. That figure is on-chip memory. The chip works out of host memory over DMA, and we’ve taken that window from 4 MiB to 768 MB per chip. Across 22 chips, that’s nearly 17 GB of addressable model memory, and it grows with every chip you add. Separately, we’ve run a 78 MiB layer bit-exact on a completely stock chip by splitting it. The same software runs on Epyc servers, NVIDIA Jetson, Qualcomm boards, and 131 cloud nodes on six continents. I’ve never positioned Akida as a large-language-model engine. My papers say the opposite. The argument is that neuromorphic processors are peers to CPUs, GPUs and quantum processors, and each kind of work belongs on the processor whose physics suits it. Placement runs on workload management capabilities already in production across thousands of sites, data centers included. On that view a Sparxity rack isn’t a rival, it’s one more tier we can work with in advancing Akida’s reach in a heterogeneous compute ontology. Both/And, not Either/Or. What Akida suits is the standing watch, among other things, and the data center has one. The costliest data movement there isn’t on a chip’s bus. It’s copying petabytes out of storage to wherever the analysis runs. So I’ve built the Engram Storage 1000: data-center storage servers with 16 to 64 Akida processors beside the data. Data is judged as it’s written and archives are re-read where they sit. Only what the fabric flags goes on to GPUs. Only the answer moves. A “hack” doesn’t come with a chain of trust or IEEE peer-reviewed design proofs. Akida serves models through vLLM, the industry’s standard serving layer, on a secure path. Every model is signed, verified bit-exact on reserved silicon before release, and checked again at use. Processors are compared with each other across the fleet, so correctness is checked continuously rather than assumed. All of it runs under the NSA’s CNSA 2.0 quantum-resistant suite, ahead of the January 2027 requirement for new national security systems. As for the comparison, read the sources, not the AI summary. SpiNNaker2 was designed for brain simulation and its LLM positioning is recent. Its installations are at universities and research centers. Jülich is a two-rack research system. The T-Systems deal is a letter of intent around a chip slated for 2028. The efficiency multiples are a homepage chart label and a single-board result with the paper still to come. The software described in its published sources is a low-level programming capability through research toolkits, reaching boards over a mere 1 Gbit Ethernet while AI data-center fabrics run at 400GbE and up. Those sources describe no signed or verified models, no post-quantum cryptography, no tenant isolation, no vLLM or standard serving endpoint, and no production workload management placing work across a fleet, and no distributed file system. That’s still a research platform, not an enterprise one. None of that is a knock on serious people doing innovative work with real hardware. It just isn’t evidence that SpiNNaker2 is somehow superior to Akida, alone or orchestrated through a heterogeneous compute ontology. “Edge” is a perspective, and so is “data center.” Neither really speaks to what’s actually needed for any given workload. The point here isn’t about hardware differences and what’s better. The point is that being different makes more work possible where it needs to run. That’s why we can say, “Run Akida anywhere. Run Akida everywhere.” And, I will keep saying it. teapawt.net/computeontology/status…
5
11
1
66
1,466
DuckMan retweeted
Labels are expensive. Here, the chip learned without cell-type labels; the host assigned names afterward using 60 labelled examples. Watch the AKD1500 learn from 8,192 unlabelled mouse-cell gene profiles, then score 9,511 cells from a second mouse. This is a continuous live screen capture of the chip run, using saved data. The host encodes the profiles, uses 60 labelled examples (four for each of 15 cell types) to name the learned memory records, and puts the predictions back on the brain map. The chip never sees positions. We repeated one earlier run: every score and measured memory state matched exactly. Agreement with atlas labels was 69.0%, balancing cell types and the three sections. This demonstrates learning and scoring on the device; it is not a new accuracy test. Conventional software methods scored higher in the earlier comparison.
1
10
37
961
DuckMan retweeted
This chip can change its learned memory right where it runs. We fed a @BrainChip_inc AKD1500 patterns from mouse brain cells, plus their known cell types. genes are instructions cells use, and different cell types use them in different amounts. that's the pattern we gave the chip. Five memory banks, trained one at a time. 6,417 lessons. every recorded response and final memory matched our software prediction exactly. We also tested a surprise from an earlier failed run: with learning enabled, unlabelled inputs can still change memory. All 80 recorded responses and 25 memory snapshots matched the prediction we filed. The video includes a live repeat using 16 saved cell profiles with no cell-type labels supplied; all 16 score sets and all 16 memory readbacks matched the earlier run exactly. A step toward devices that adapt locally. The computer still prepares inputs, provides labels and combines the banks. Accuracy was measured in software, not on chip.
3
10
2
50
1,979
DuckMan retweeted
Yes. Hardware efficiency gains of ~30-40%/yr, sparse models, and advancing neuromorphic designs compound hard. Continuous AGI-level inference at 100W (roughly 5× the human brain) looks plausible by end of 2036. Early systems will start hungrier, then compress toward that.
1
3
78
DuckMan retweeted
TRENDING: Semiconductor technology development roadmap is to combine chiplet-based architectures, domain-specific silicon, and edge AI processors, with longer-term bets on neuromorphic computing and silicon photonics.
1
4
77
Neuromorphic Quantum Machine Learning.
what’s the next big thing after AI?
1
1
216
DuckMan retweeted
How do we build smarter, more energy-efficient AI? Through THOR: The Neuromorphic Commons, the nation's first open-access neuromorphic computing hub, researchers are gaining unprecedented access to the tools, infrastructure and partnerships needed to push brain-inspired AI forward. It's a pioneering effort backed by a $4 million @NSF investment. Learn more and see today's piece in The Chronicle of Higher Education: ai.utsa.edu. @AI_UTSA | @UTSAResearch
1
1
6
1,041
Replying to @bottegamd
Neuromorphic Ai , will usher in much needed change one of these days
1
2
257
SGLang solves KV cache scheduling, but Von Neumann physics remains the wall. Shuttling bits between HBM and tensor cores burns 80% of cluster power. Neuromorphic co-locates memory and synaptic compute. Until transformers embrace event-driven sparsity, power caps cluster scale.
1
3
178