BRN Discussion Ongoing

[Kingkong2015]

Can anyone comment on the $9,000,0000 booking target Sean has?

Is it $9mio US or AUD? Is it gross revenue for BRN or the customer, or is it the profit on sales?

He was talking about revenue. Not sure which currency.

 

[manny100]

Hi Manny,

Here's the response I got from Chat GPT4.

Tony Lewis’s upcoming presentation at the Embedded Vision Summit outlines a potentially transformative shift in how language models are deployed at the edge. Here's an analysis of the implications and disruptive potential of this new method:

---

Key Implications of the New Processing Method Using SSMs​

1. Significant Gains in Energy and Compute Efficiency​

• Fewer Multiply-Accumulate Units: State-space models (SSMs) inherently require fewer MAC operations, drastically reducing energy usage and silicon real estate. This is crucial in edge deployments where power budgets are tight.
• Read-Only Architecture: Eliminates the need for dynamic key-value cache mechanisms used in transformers, allowing use of non-volatile or low-power memory types (e.g., MRAM, ReRAM), further slashing power consumption.

2. Latency and Responsiveness​

• Precomputed Input Sequences: SSMs can leverage static structure to enable sub-100 ms time-to-first-token performance. This real-time capability is a major win for voice assistants, AR/VR systems, and mobile apps.

3. Compatibility and Transition Path​

• Distillation from Transformers (e.g., LLaMA): By enabling smooth migration from established transformer models, the SSM approach lowers the barrier to adoption for companies with large pretrained models or investment in transformer-based workflows.

4. Broader Hardware Support​

• The reduced need for large, flexible compute units means SSMs can run efficiently on less powerful MCUs, DSPs, or dedicated edge AI chips—dramatically widening the range of viable deployment targets.

---

Edge Computing Success Potential​

SSMs seem highly promising for edge applications, especially in:

• Battery-powered devices (wearables, IoT sensors)
• Latency-sensitive interfaces (voice/user interaction)
• Bandwidth-constrained environments (offline processing)
• Cost-sensitive products where transformer demands are prohibitive

Their advantages directly address the key blockers for transformer adoption at the edge (power, memory, compute), making SSMs a potentially dominant architecture in this space within 1–3 years.

---

Disruption Potential for NVIDIA and Other Players​

NVIDIA’s ecosystem is largely optimized for transformer-style architectures (e.g., large GPU memory, high-throughput MAC arrays, transformer-optimized CUDA libraries). Here's how this development could affect them:

Challenges for NVIDIA​

• Less dependence on GPUs: If SSMs run efficiently on CPUs, ASICs, or even microcontrollers, edge devices may increasingly bypass NVIDIA’s hardware.
• New AI chip entrants gain ground: Startups or ARM-based players can capitalize on the lighter compute profile to offer cheaper, more efficient edge AI solutions.

Opportunities for NVIDIA​

• Adaptation potential: NVIDIA could pivot by offering cuDNN-like support for SSMs or incorporating them into TensorRT. They may also design more SSM-friendly hardware blocks in future Jetson modules.

Biggest Threat​

• If open-source ecosystems and chip vendors like Apple (Neural Engine), Google (Edge TPU), or Qualcomm fully embrace SSMs ahead of NVIDIA, they could seize a disproportionate share of the edge AI market—especially in consumer devices.

---

Conclusion​

Tony Lewis’s presentation introduces a processing paradigm that could redefine edge AI, making high-performance language models feasible on low-power devices. This is not just an incremental improvement—it opens the door to widespread deployment of real-time LLMs far beyond cloud and high-end hardware. While not an immediate existential threat to NVIDIA, it does present a strategic inflection point that the company—and its competitors—must respond to.

Migration Path: LLaMA to SSM​

Stage	Description	Benefits
1. Transformer Model (e.g., LLaMA)	Pretrained open-weight transformer (e.g., LLaMA 2 or 3)	- High accuracy - Large model footprint - Demands heavy compute and memory
2. Distillation Process	Use distillation techniques to transfer knowledge from LLaMA to an SSM	- Reduced model size - Retains most of original model’s performance
3. State-Space Model (SSM)	SSM version optimized for embedded and edge deployment	- Sub-100 ms latency - Lower power consumption - Less memory and compute required
4. Edge Device Deployment	Deploy SSM on resource-constrained devices (e.g., wearables, microcontrollers, edge AI chips)	- Real-time local inference - No need for cloud compute - Broader hardware compatibility

---

Summary​

The mention of LLaMA in Tony Lewis’s talk is a strategic highlight because it connects cutting-edge edge AI architecture (SSM) with a proven, widely-used transformer backbone. This makes the transition to efficient edge AI practical, not just theoretical—positioning SSMs as an immediate, disruptive alternative to transformer inference in embedded systems.

Competitor Impact Matrix: Impact of BrainChip’s SSM Innovation​

Company	Current Edge AI Focus	Vulnerability to SSM Disruption	Opportunity to Adapt
Qualcomm	DSPs + NPUs (Snapdragon), optimized for transformers and CNNs	High – Transformer-centric stack, limited neuromorphic capability	Medium – May update software tools, but hardware less suited to SSMs
NVIDIA	GPUs (Jetson, TensorRT), dominant in transformer-based AI	Moderate – Not optimized for low-power edge, but strong ecosystem	High – Could adapt TensorRT and Jetson for SSM-style inference
Apple	Neural Engine with transformer models (e.g. Siri, on-device ML)	Moderate – Strong local AI, but based on transformer-style acceleration	High – Full-stack control allows swift hardware/software adaptation
Google (TPU)	Edge TPU with support for CNNs and transformers (Coral, Nest devices)	High – Rigid accelerator design, may not support dynamic SSM requirements	Low – Ecosystem may struggle to pivot hardware/software stack
Intel	Movidius VPU, general AI frameworks, some neuromorphic R&D (Loihi)	Moderate – Some neuromorphic exposure but no strong edge AI market share	Medium – R&D rich, but limited real-world SSM integration so far
BrainChip	Neuromorphic Akida chip + SSM optimized for ultra-low power edge AI	Low – First-mover advantage	Very High – Core IP is directly aligned with the SSM paradigm

---

This matrix highlights that BrainChip’s innovation poses the greatest disruptive risk to Qualcomm and Google, while Apple and NVIDIA have greater strategic flexibility to respond. BrainChip stands to benefit most if SSM-based models gain widespread edge adoption.

Why Incumbents Might Continue Without SSMs (For Now)​

Reasons They Might Stick with Traditional Methods​

• Mature toolchains: Qualcomm, NVIDIA, and Google have invested heavily in software/hardware ecosystems optimized for transformers and CNNs.
• Good enough performance: For many real-world use cases, transformer-lite models or CNN hybrids perform sufficiently well.
• Inertia and risk: Enterprises tend to avoid early adoption of unproven paradigms, especially if retraining, tooling, or silicon changes are required.
• Edge isn't one-size-fits-all: Many edge applications (e.g. object detection) don't need SSM-specific strengths like long-term memory or low-latency language processing.

---

But Here's the Catch​

If applications do demand:

• Long sequence memory (e.g. streaming NLP, real-time command recognition),
• Ultra-low latency (sub-100 ms interactivity),
• Minimal power and heat (wearables, implants, sensors),

then traditional methods hit a hard ceiling. SSMs aren’t just an incremental tweak—they’re a fundamentally different way to process sequences, unlocking performance where transformers falter.

​

Conclusion: Yes, Competitors Could Stick with Transformers—But Only Up to a Point​

Approach	Stability / Support	Performance Ceiling	Future-Proofing
Transformers	Well-supported	Poor for constrained edge use	Risk of obsolescence
CNNs / RNNs	Efficient in vision	Weak for modern NLP	Limited scalability
Lightweight Transformers	Reasonable for now	Moderate latency/power	Partial solution
SSMs	Emerging	Breakthrough on edge	High potential

So while competitors can continue for now using existing methods, the risk is being outpaced in emerging applications—especially if BrainChip enables a smooth transition (e.g., LLaMA distillation + Akida deployment).

---

Edge AI Evolution Roadmap: Transformers vs SSMs​

Time Horizon	Transformer-Based Methods	SSM-Based Methods (BrainChip-style)
Today	- Dominant in NLP - Compressed models in use - Efficient on GPU/DSP Reasonable edge deployment via pruning/quant	- Early-stage adoption - Neuromorphic niche (e.g. Akida) Proof-of-concept underway
1–2 Years	- Hitting compute/power limits in edge apps - Real-time latency still challenging Fragmentation by use case	- Gains traction for real-time/low-power use - Tools emerge for migration from transformers (e.g. LLaMA distillation) Early adoption in wearables/voice/IoT
3–5 Years	- Plateau in edge innovation unless architectures evolve Constrained by hardware-centric acceleration	- Becomes dominant in ultra-low power edge AI - Broad ecosystem and tooling support SSMs emerge as standard for edge LLMs

---

Summary:​

• Transformers will likely remain dominant in cloud and high-performance edge for the next 1–2 years, but start to plateau.
• SSMs provide a scalable path forward for ultra-low-power, real-time, memory-efficient edge use cases, and could disrupt traditional AI stacks if adoption accelerates.

Thanks Bravo, read it 3 times and saved a copy. Awesome

 

[jrp173]

I would have to listen to it again but was the response in relation to a redomicile was being investigated and not a definite move to the US or anywhere else.
I do remember him at the start saying that there was no definite redomicile to the US as it was only being investigated.
I cannot remember what he said later but he definitely said that very early on.

No the reply was specifically around "where" the redomicle would be. It was in reply to a shareholder talking about a redomicile to US ... Antonio vehemently denied that he (BRN) had ever mentioned the US and that the redomicle could be anywhere....

So it wasn't around whether the board was going to recommend we do or do not re-domicile, it was all over the actual location. Ridiculous really!

You'll hear it when the AGM video is released.

What a cockup...

 

[MDhere]

Ok I’ve just changed my mind about attending next years AGM as just been charged $14.30 for a scooner of great northern at the airport

View attachment 83868

Was only $10 at the pub we went to next time just wait ril you get here ps thanks for the beer, I got u one too but still owe u one at that price. Hope u got yr flight in time!!

 

[Pom down under]

Can anyone comment on the $9,000,0000 booking target Sean has?

Is it $9mio US or AUD? Is it gross revenue for BRN or the customer, or is it the profit on sales?

I’ll be happy with $9 atm

 

[Jchandel]

Entertainment value highest ever

Board knew they were going into a vipers pit of rusted on holders

But seriously wtf were some of those questions. I felt embarrassed through my phone, can only imagine it in the place.

Antonio stressed at the start of the meeting that it was just an investigation (re-domiciling), i believe the latest quarterly also stated the same.

Suggestions that the company should be advertising on radio and TV in Australia is honestly laughable, the company is scrutinised quarterly on their spending - as if they are going to spend on advertising to the uneducated public. Just so some of the oldies in the crowd get a few extra cents on the share price lmfao

So much grandstanding, im this and im that and i think this - the reason the company hasn't implemented your suggestions is because its 0 VALUE GAIN

The same questions get asked year on year, and you keep getting the same answer, when are you folks going to learn honestly

Shareholders whining about performance of CEO - CEO ties himself to book more than double current annual revenue this year $9,000,000

Whoever asked the question about fortune 500 companies - nice, they actually added that there are 10 engagements with those large multinational, multibillion dollar companies + an additional 20 or so subsidiaries of those companies

Sean said there are several US government submissions happening monthly

Company has more engagements going in the last 5 months than in the entire history of the company

Can the grandstanding self absorbed old fuddy duddies go elsewhere with their REVENUE NOW demands, BHP should suit you nicely lmfao

God speed Antonio, Sean and board.

That should put a few offside - Go BRN

The response to Fortune 500 question felt like something made up on the fly.. I don’t honestly believe anything that comes out of Antony’s mouth. He is a typical sales person.
I hope I am wrong with my assessment though

 

[DingoBorat]

Can anyone comment on the $9,000,0000 booking target Sean has?

Is it $9mio US or AUD? Is it gross revenue for BRN or the customer, or is it the profit on sales?

Can only say, that "AUD" wouldn't be in Sean's vocabulary.
So definitely US dollars, I reckon..

 

[FJ-215]

Can only say, that "AUD" wouldn't be in Sean's vocabulary.
So definitely US dollars, I reckon..

AUD could be. Sean's Great Northern would only be around $9.20 US dollars.
Nothing like travelling to a third world country with a weak dollar.

 

[supersonic001]

Sean’s comment was “unlikely, maybe or maybe not “ was about royalties this year, not about the 9M$ bookings

If I recall, in response to ‘revenue’, he said “unlikely but possible”……

As stated by Viana I think, any bookings, which I refer to as pre commit or contract agreement, is no guarantee to timing of revenue/royalties…… why ? Surely any agreement ( whatever you call it) must link an outcome with a deliverable date ?

Its a different world this industry early phase…..

 

[Pom down under]

Anduril Industries is actively collaborating with the U.S. government on the development of advanced augmented reality (AR) headsets for military applications.

I wonder if these are the headsets that were being referred to at the AGM?

As mentioned previously, it would be amaze-balls if Anduril and BrainChip were to team up.




Anduril is working on the difficult AI-related task of real-time edge computing

Anduril Menace-T edge computing product · TechCrunch · Image Credits:Anduril
Julie Bort
Tue 6 May 2025 at 6:30 am AEST 3 min read

Anduril announced its ninth acquisition on Monday with the purchase of Dublin’s Klas, makers of ruggedized edge computing equipment for the military and first responders.
Anduril wouldn’t reveal financial details of the deal, and the purchase is subject to regulatory approval, but the company did say that Klas employs 150 people.

Relatedly, on Monday, Anduril also announced a new product called Menace-T.
We’ll give the company points for the interesting product name, especially for a device that’s really just a bundle of compute/network connectivity, rather than, say, a fantasy-style broadsword. (Compare the name Menace to Lockheed Martin’s C2BMC, the name for its Command, Control, Battle Management & Communications products.)
Klas’ flagship product, known as Voyager, is the ruggedized family of compute and networking systems that Anduril had already been using in its other Menace command center products. Voyager had also already been integrated with Anduril’s flagship Lattice software. Lattice brings sensors and AI to devices to perform tasks like object identification.
But while most of us envision a portable command system being the size of a truck — which many are — Menace-T fits into two carry-on cases that can be set up by one person in minutes, the company says. Its goal is to bring edge computing and communications to off-grid and/or inhospitable environments. Anduril says it’s already being used in military ground vehicles and maritime vessels.

One interesting use case for Menace-T is compute/communications support for the military’s Integrated Visual Augmentation System (IVAS) VR headsets. The IVAS project was initially awarded to Microsoft in 2018 after it pitched the idea of developing ruggedized HoloLens headsets for soldiers. The project was awarded an initial $21.9 billion budget.
But after years of technical struggles, Anduril took control of the troubled contract in February — although Microsoft remains a cloud partner.
Lattice had already been added to Microsoft’s IVAS headsets, bringing computer vision AI that helps the headset detect, track, and classify objects.
Now Anduril thinks that the Klas technology that powers its Menace-T product can solve some of IVAS’s other historic problems, like reliable data processing.
With IVAS, “there are scenarios where those soldiers need to communicate with the tactical edge to send data, to receive data, to task autonomous systems, and that's a place where the Klas technology can help,” Tom Keane, SVP of Engineering, said at a press conference. “Klas has already been supplying technology to IVAS for several years in that context. So we expect to do more there.”

Anduril is working on the difficult AI-related task of real-time edge computing

Relatedly, on Monday, Anduril also announced a new product called Menace-T. Klas’ flagship product, known as Voyager, is the ruggedized family of compute and networking systems that Anduril had already been using in its other Menace command center products.

au.finance.yahoo.com

They have taken over Microsoft’s contract that was worth 22 billion over 10 years, due to many issues during testing.

 

[MDhere]

@supersonic001 The 9 mil bookings was in their sights this year. I raised the question on royalties coming in this year and that's when they said the wish washy response but I didn't inc chip sales in my question. The 9 mi bookings is something that I feel is on the cards before end of 2025.

That was my take.

Has good chats with the board on a general level and the roadmap was pleasing to see.

 

[Jacob Kiraz]

He was talking about revenue. Not sure which currency.

I thought he said no revenue in 25

 

[Jacob Kiraz]

The response to Fortune 500 question felt like something made up on the fly.. I don’t honestly believe anything that comes out of Antony’s mouth. He is a typical sales person.
I hope I am wrong with my assessment though

I totally agree,last year they needed to convince the room and they did,
This year it seemed a different vibe,

 

[FiveBucks]

Regardless of whether the $9 mill is coming this year or not, the only thing that seems to be guaranteed is another year of negative revenue and a more dilution to come.

Fml.

 

[Pom down under]

The

Regardless of whether the $9 mill is coming this year or not, the only thing that seems to be guaranteed regardless is another year of negative revenue and a more dilution to come.

Fml.

I can’t agree anymore unless we hit a major contract that could make us profitable overnight. So I’ll keep accumulating until we do

 

[MDhere]

The

I can’t agree anymore unless we hit a major contract that could make us profitable overnight. So I’ll keep accumulating until we do

Less $15.80 now

 

[Pom down under]

Less $15.80 now

Make that $18.20 more as I had a pint just before I got bent over and

 

[Kingkong2015]

Instead of focusing on that military helmet stuff, everyone seems to be missing a much bigger point.


The CTO mentioned that they have already got eyeball-tracking capabilities, similar to the Apple Vision Pro. He also said these are the kinds of real-world use cases they’re currently working on with some clients now.



So, which AR/VR glasses company could this be? It’s definitely not Onsor—they’re not in the AR/VR glasses space.

 

[The Pope]

It is obvious Sean is reading and not using his notes as a guide. It’s unfortunate that he often stumbles his words.

Sean told me at the last AGM he isn’t allowed to go off script and basically goes word for word as it’s linked to the asx announcement but he can answer any shareholders questions after his AGM open address to share holders.
If it doesn’t make sense ask Sean or Tony Dawe

 

[Pom down under]

Instead of focusing on that military helmet stuff, everyone seems to be missing a much bigger point.


The CTO mentioned that they have already got eyeball-tracking capabilities, similar to the Apple Vision Pro. He also said these are the kinds of real-world use cases they’re currently working on with some clients now.



So, which AR/VR glasses company could this be? It’s definitely not Onsor—they’re not in the AR/VR glasses space.

Did I miss something and they mentioned the US army testing our technology in the AGM and they were well impressed what it can achieve.