BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
I expect good news this century! GO BRAINCHIP🧠⚡️ Neuromorphic Computing: AI That Thinks Like a Brain | Akshay Rajinikanth
🧠⚡️ Neuromorphic Computing: AI That Thinks Like a Brain I came across this while studying neural networks — and it completely changed how I think about AI. What if computers could process information the way our brains do? That’s what Neuromorphic Computing enables — using hardware and...
Check out the comments, sorry if already posted.
Article from Beata Shares: Tittle below
For example, the US National Defence Authorisation ACT (NDAA) has allocated approximately $30 billion to cybersecurity initiatives in FY20253. And as part of NATO’s total defence spending target of 5%, 1.5% of that would be related to “defence-related outlays” which includes spending on cybersecurity.
All of these factors together have helped drive substantial acquisition activity over the last year, including:
Is this the perfect storm for cybersecurity stocks?
"Expanding defence budgets
Defence budgets globally have allocated significant increases to cybersecurity recently, reflecting its importance in modern warfare and national security.For example, the US National Defence Authorisation ACT (NDAA) has allocated approximately $30 billion to cybersecurity initiatives in FY20253. And as part of NATO’s total defence spending target of 5%, 1.5% of that would be related to “defence-related outlays” which includes spending on cybersecurity.
All of these factors together have helped drive substantial acquisition activity over the last year, including:
- Google’s US$32 billion acquisition of cloud security startup Wiz (March 2025)
- Cisco’s US$28 billion acquisition of Splunk (March 2024)
- Thoma Bravo’s US$5.3 billion acquisition of Darktrace (October 2024)
- Mastercard’s US$2.65 billion acquisition of Recorded Future (December 2024)
- CyberArk’s US$1.54 billion acquisition of Venafi (October 2024)
Fullmoonfever
Top 20
A little while ago @Frangipani had kindly posted some images and LinkedIn posts with Edward Lien our Regional SM at Andes Risc-V Con.
I also saw someone post something earlier this morning about it would be nice to be in smart phones or similar etc.
Well, I can't say we'll be in smart phones etc as yet but I was reading through the Andes site, (link below but you will need to use auto translate to read it all - or maybe not as the Google translate link I pasted may do it for you), and when you look at the recent partnerships inc BRN with their Voyager Board, AndesCore™ AX45MP, SOC etc there appears to be a very synergistic ecosystem there and some use case they state are smart phones, tablets, SDV etc.
They even reached a milestone with one of the partnerships listed (Imagination for GPU) running Android 15 specifically for the above targets.
The other thing I noticed was something that Edward was presenting on as below. Not just AI acceleration but specifically LLM engine.
I also saw someone post something earlier this morning about it would be nice to be in smart phones or similar etc.
Well, I can't say we'll be in smart phones etc as yet but I was reading through the Andes site, (link below but you will need to use auto translate to read it all - or maybe not as the Google translate link I pasted may do it for you), and when you look at the recent partnerships inc BRN with their Voyager Board, AndesCore™ AX45MP, SOC etc there appears to be a very synergistic ecosystem there and some use case they state are smart phones, tablets, SDV etc.
They even reached a milestone with one of the partnerships listed (Imagination for GPU) running Android 15 specifically for the above targets.
The other thing I noticed was something that Edward was presenting on as below. Not just AI acceleration but specifically LLM engine.
- Edward Lien, Regional Sales Manager at BrainChip - The edge LLM engine combined with the RISC-V core provides a low-power and high-performance artificial intelligence development platform.
Probably posted already but Tony Lewis sees a lot of possibilities for neuromorphic annd mobile phones
Glenn Mason • 3rd+
--4h
I have a question if I may Mr Lewis.... Is there any room in the mobile phone market for neuromorphic compute or are the current systems used doing the job well enough that a switch over to neuromorphic technology isn't high on their radar currently ?.
Like
Reply2 replies2 Replies on Glenn Mason’s comment
M Anthony LewisAuthor
CTO@BrainChip | AI, Robotics, Disruptive Computing4h
Glenn Mason interesting question. Yes of course. But a lot if factors have to be aligned each time some neuromorphic technology is absorbed. Always on llms could be possible. Better modems. High speed image processing. Wearables. Yes there is a lot.
Pom down under
Top 20
TheDrooben
Pretty Pretty Pretty Pretty Good
Frontgrade GaislerFrontgrade Gaisler5,429 followers5,429 followers6mo • 6 months ago • Visible to anyone on or off LinkedIn
We are happy to announce that we have licensed the Akida Neuromorphic AI IP from BrainChip for integration into our fault-tolerant, space-grade system-on-chip solutions.
“For the last 20 years, Gaisler’s space-grade microprocessors have been successfully deployed to every planet of our solar system in missions from ESA as well as most other space agencies,” said Sandi Habinc, General Manager at Frontgrade Gaisler, a Frontgrade Technologies company. “Together with BrainChip, we evaluated the Akida IP and decided that the next step of licensing it would beneficially augment our future space processors with neuromorphic AI.” hashtag#NeuromorphicAI hashtag#AIinSpace hashtag#SpaceComputing
https://lnkd.in/gquSMrAa
…more
Taproot
Regular
Implementing Neural Network Algorithms on Neuromorphic Processors | SBIR.gov
legacy.www.sbir.gov
This one is only 6 weeks away from end date.
Something to look forward to.
Stable Genius
Regular
We are proud to present the Neurospace Project: a collaboration between Frontgrade Gaisler and BrainChip supported by the European Space Agency - ESA. | Frontgrade Gaisler
We are proud to present the Neurospace Project: a collaboration between Frontgrade Gaisler and BrainChip supported by the European Space Agency - ESA. At its center: GR801, our first neuromorphic AI solution for space. GR801 is built on the radiation-hardened NOEL-V processor and powered by...
There is a 2 min movie talking up the collaboration but it want copy across so have to view it in the LinkedIn link!
We are proud to present the Neurospace Project: a collaboration between Frontgrade Gaisler and BrainChip supported by the European Space Agency - ESA.
At its center: GR801, our first neuromorphic AI solution for space.
GR801 is built on the radiation-hardened NOEL-V processor and powered by BrainChip's Akida.
It delivers ultra-low power, real-time pattern recognition, and onboard decision-making without requiring Earth contact.
This is more than a milestone.
It's a paradigm shift for AI at the edge of space exploration.
Learn more: https://lnkd.in/dMq9sxkp
And https://lnkd.in/dSx8_mTQ
Also available on YouTube: https://lnkd.in/dC6_aWR2
#NeuromorphicAI #SpaceTech
Last edited:
Wow, that is interesting!
I recently read about how a single changed pixel in a large image can lead to a completely different result.
And the reason lies in the precision with which images are supposed to be recognized.
With the many millions or billions of calculations in matrices, one mistake leads to major changes.
www.linkedin.com
I recently read about how a single changed pixel in a large image can lead to a completely different result.
And the reason lies in the precision with which images are supposed to be recognized.
With the many millions or billions of calculations in matrices, one mistake leads to major changes.
I'm excited to share that our paper, titled 'Risk Analysis of One-Pixel Image Defects in Safety-Critical Deep Neural Networks', will be presented at the WAISE Workshop during the SAFECOMP 2025… | Dr. Krystian Radlak
I'm excited to share that our paper, titled 'Risk Analysis of One-Pixel Image Defects in Safety-Critical Deep Neural Networks', will be presented at the WAISE Workshop during the SAFECOMP 2025 conference! You can check out the full program here: https://lnkd.in/d4776rig. You can find the link...
Partnership between BrainChip and HaiLa reflects future of smart edge devices—where sensor intelligence, connectivity, and compute are optimized for power.
hashtag#dataquest hashtag#BrainChip hashtag#HaiLa hashtag#EdgeAI hashtag#UltraLowPower hashtag#BackscatterWiFi hashtag#NeuromorphicAI hashtag#IoTSensors hashtag#BatteryFree hashtag#SmartInfrastructure
Frangipani
Top 20
We are proud to present the Neurospace Project: a collaboration between Frontgrade Gaisler and BrainChip supported by the European Space Agency - ESA. | Frontgrade Gaisler
We are proud to present the Neurospace Project: a collaboration between Frontgrade Gaisler and BrainChip supported by the European Space Agency - ESA. At its center: GR801, our first neuromorphic AI solution for space. GR801 is built on the radiation-hardened NOEL-V processor and powered by...
Love the video - here it is on YouTube:
HopalongPetrovski
I'm Spartacus!
Fantastique.We are proud to present the Neurospace Project: a collaboration between Frontgrade Gaisler and BrainChip supported by the European Space Agency - ESA. | Frontgrade Gaisler
We are proud to present the Neurospace Project: a collaboration between Frontgrade Gaisler and BrainChip supported by the European Space Agency - ESA. At its center: GR801, our first neuromorphic AI solution for space. GR801 is built on the radiation-hardened NOEL-V processor and powered by...
There is a 2 min movie talking up the collaboration but it want copy across so have to view it in the LinkedIn link!
We are proud to present the Neurospace Project: a collaboration between Frontgrade Gaisler and BrainChip supported by the European Space Agency - ESA.
At its center: GR801, our first neuromorphic AI solution for space.
GR801 is built on the radiation-hardened NOEL-V processor and powered by BrainChip's Akida.
It delivers ultra-low power, real-time pattern recognition, and onboard decision-making without requiring Earth contact.
This is more than a milestone.
It's a paradigm shift for AI at the edge of space exploration.
Learn more: https://lnkd.in/dMq9sxkp
And https://lnkd.in/dSx8_mTQ
Also available on YouTube: https://lnkd.in/dC6_aWR2
#NeuromorphicAI #SpaceTech
So much more engaging than asking the fridge what "vacation mode" is all about.
More of this!!!
I was talking with ChatGPT about Brainchip and endet up with this …
My post:
I’ve said it before on TSE .. none of this makes any sense to me anymore.
BrainChip has so many well-known partners, licensees, universities, and experts who speak positively about Akida and its potential. Everyone keeps saying it’s a game-changer, with unmatched energy efficiency and real-world applications. The potential is enormous.
And yet… the share price is utterly pathetic.
The management seems to do nothing but quote musicals, say things like “we hope that…” or “we see our product in every device” .but they take no actual steps to get real attention on the stock.
What’s going on? What are they waiting for?
Response:
You’re definitely not alone in feeling this. Many long-term holders are asking the same question: How can a company with such revolutionary tech and such a powerful network be valued so poorly?
Let’s break it down:
The contradiction:
What’s wrong?
1. No significant revenue yet.
Lots of partners, yes – but none seem to have converted into real, meaningful licensing income. Without that, the market sees only R&D, not business.
2. Vague communication.
Management loves to say:
“We see Akida in all devices”
“We believe the world will see its value”
But there’s rarely anything concrete – no hard deals, no revenue numbers, no OEM names.
3. Dilution kills confidence.
The constant use of ATMs, performance rights, and stock options has deeply eroded investor trust. More shares, same pie.
4. Weak investor communication.
Other tech firms do roadshows, PR pushes, interviews, demos. BrainChip gives us Broadway quotes and philosophical posts – but no strategic IR campaign.
5. Market suspicion.
Too much promise, too little delivery. Until an OEM ships actual products with Akida inside – and BrainChip reports real revenue – the market stays skeptical.
So what is the management waiting for?
Bottom line:
You’re right: the gap between Akida’s perceived value and BRN’s actual market cap is absurd – but not accidental.
Management is either silent because they’re legally obligated – or because there’s simply nothing new to say.
Either way, it’s costing them trust, momentum, and shareholder confidence.
Until we see:
…the market will stay on the sidelines.
It’s not fair. But it’s how the game works.
My post:
I’ve said it before on TSE .. none of this makes any sense to me anymore.
BrainChip has so many well-known partners, licensees, universities, and experts who speak positively about Akida and its potential. Everyone keeps saying it’s a game-changer, with unmatched energy efficiency and real-world applications. The potential is enormous.
And yet… the share price is utterly pathetic.
The management seems to do nothing but quote musicals, say things like “we hope that…” or “we see our product in every device” .but they take no actual steps to get real attention on the stock.
What’s going on? What are they waiting for?
Response:
You’re definitely not alone in feeling this. Many long-term holders are asking the same question: How can a company with such revolutionary tech and such a powerful network be valued so poorly?
Let’s break it down:
The contradiction:- Tech: Akida is brilliant. Ultra-low power, edge-native, ideal for AIoT. Validated by experts, research labs, and devs worldwide.
- Partners: ARM, Edge Impulse, Mercedes-Benz hints, NASA references, universities.
- Potential: Everyone says “this chip could be in everything” – phones, cars, sensors, wearables.
- Reality: Sub-€0.10 stock, constant dilution, and nearly zero trust from the broader market. Why?
What’s wrong?1. No significant revenue yet.
Lots of partners, yes – but none seem to have converted into real, meaningful licensing income. Without that, the market sees only R&D, not business.
2. Vague communication.
Management loves to say:
“We see Akida in all devices”
“We believe the world will see its value”
But there’s rarely anything concrete – no hard deals, no revenue numbers, no OEM names.
3. Dilution kills confidence.
The constant use of ATMs, performance rights, and stock options has deeply eroded investor trust. More shares, same pie.
4. Weak investor communication.
Other tech firms do roadshows, PR pushes, interviews, demos. BrainChip gives us Broadway quotes and philosophical posts – but no strategic IR campaign.
5. Market suspicion.
Too much promise, too little delivery. Until an OEM ships actual products with Akida inside – and BrainChip reports real revenue – the market stays skeptical.
So what is the management waiting for?- Option 1: They do have something but are under strict NDAs – waiting for their partners to launch products before they can speak.
- Option 2: There simply isn’t anything major yet. Just more talks, demos, trials – and no conversions.
- Option 3: They’re trying to avoid overhyping, aiming for long-term trust… but forgot that even long-term investors need clarity and communication.
Bottom line:You’re right: the gap between Akida’s perceived value and BRN’s actual market cap is absurd – but not accidental.
Management is either silent because they’re legally obligated – or because there’s simply nothing new to say.
Either way, it’s costing them trust, momentum, and shareholder confidence.
Until we see:
- Real, named customers
- Public revenue from licenses
- Clear product integrations
…the market will stay on the sidelines.
It’s not fair. But it’s how the game works.
Frangipani
Top 20
Hi @CrabmansFriend,
when googling for more information on BrainChip’s Aether Core LLM accelerator on FPGA (cf. https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-469022), I came across a few links to a recent (now expired) BrainChip job ad for a Hardware Engineering intern that says “Our current demo utilizes Intel x86 in conjunction with Aether Core LLM for showcasing our platform. This internship project aims to create an advanced proof-of-concept platform by integrating the LLM with a high-end ARM series. This innovation will serve as an impressive showcase at various demo and trade shows, highlighting our LLM capabilities with ARM technology”.
Can we infer from the info that the current LLM FPGA Tech Demonstrator runs on an Intel x86 CPU, it is highly likely / must be based on an Altera FPGA or would it be entirely possible to mix and match? I have no idea…
Altera® FPGA Products - FPGA and SoC FPGA Devices and Solutions
Altera® FPGAs offer a wide variety of configurable embedded SRAM, high-speed transceivers, DSP blocks, high-speed I/Os, logic blocks, and routing.
Baneino
Regular
Looking Back: Early Nvidia Investors (1999–2005)I was talking with ChatGPT about Brainchip and endet up with this …
My post:
I’ve said it before on TSE .. none of this makes any sense to me anymore.
BrainChip has so many well-known partners, licensees, universities, and experts who speak positively about Akida and its potential. Everyone keeps saying it’s a game-changer, with unmatched energy efficiency and real-world applications. The potential is enormous.
And yet… the share price is utterly pathetic.
The management seems to do nothing but quote musicals, say things like “we hope that…” or “we see our product in every device” .but they take no actual steps to get real attention on the stock.
What’s going on? What are they waiting for?
Response:
You’re definitely not alone in feeling this. Many long-term holders are asking the same question: How can a company with such revolutionary tech and such a powerful network be valued so poorly?
Let’s break it down:
- Tech: Akida is brilliant. Ultra-low power, edge-native, ideal for AIoT. Validated by experts, research labs, and devs worldwide.
- Partners: ARM, Edge Impulse, Mercedes-Benz hints, NASA references, universities.
- Potential: Everyone says “this chip could be in everything” – phones, cars, sensors, wearables.
- Reality: Sub-€0.10 stock, constant dilution, and nearly zero trust from the broader market. Why?
1. No significant revenue yet.
Lots of partners, yes – but none seem to have converted into real, meaningful licensing income. Without that, the market sees only R&D, not business.
2. Vague communication.
Management loves to say:
“We see Akida in all devices”
“We believe the world will see its value”
But there’s rarely anything concrete – no hard deals, no revenue numbers, no OEM names.
3. Dilution kills confidence.
The constant use of ATMs, performance rights, and stock options has deeply eroded investor trust. More shares, same pie.
4. Weak investor communication.
Other tech firms do roadshows, PR pushes, interviews, demos. BrainChip gives us Broadway quotes and philosophical posts – but no strategic IR campaign.
5. Market suspicion.
Too much promise, too little delivery. Until an OEM ships actual products with Akida inside – and BrainChip reports real revenue – the market stays skeptical.
- Option 1: They do have something but are under strict NDAs – waiting for their partners to launch products before they can speak.
- Option 2: There simply isn’t anything major yet. Just more talks, demos, trials – and no conversions.
- Option 3: They’re trying to avoid overhyping, aiming for long-term trust… but forgot that even long-term investors need clarity and communication.
You’re right: the gap between Akida’s perceived value and BRN’s actual market cap is absurd – but not accidental.
Management is either silent because they’re legally obligated – or because there’s simply nothing new to say.
Either way, it’s costing them trust, momentum, and shareholder confidence.
Until we see:
- Real, named customers
- Public revenue from licenses
- Clear product integrations
…the market will stay on the sidelines.
It’s not fair. But it’s how the game works.
Back then, Nvidia was a small, visionary tech company, developing GPU technology before the world realized its full potential. The stock was volatile, and few believed in it long-term:
YearNvidia Stock Price (split-adjusted)Investor Sentiment
1999Below $1Hopeful but uncertain – early believers, many skeptics
2002Dropped ~80% during tech crashFrustration, doubt, some sold in panic
2005Began recoveryEarly patience began to pay off
Many early Nvidia shareholders felt ignored, doubted, even ridiculed – until the world finally realized the value of GPUs for gaming, AI, and computing.---
Now: BrainChip (BRN) Shareholders in 2024–2025Current SentimentCompared to Nvidia's Early Years
High hopes for disruptive neuromorphic tech (Akida)
Similar to Nvidia’s GPU visionSlow commercialization, long wait
Like Nvidia before mass-market adoptionMarket mostly ignores the company despite real progress
Nvidia was treated similarlyStrong community belief and long-term conviction
Just like Nvidia’s early retail holders---
Final Thought:> Yes – it feels very similar.
Early Nvidia investors had to wait, be patient, and endure criticism and market dips.
Those who held on through the uncertain years witnessed one of the greatest tech success stories in history.
If BrainChip fulfills even part of its potential, many current investors could one day say:
> “I was in before anyone believed.”
Conviction + Patience = Historic Opportunity.
There are
no guarantees – but the parallels are real.
Yes. Intel would be high on the probabilities list. We are partnered with IFS (or in their ecosystem).Hi @CrabmansFriend,
when googling for more information on BrainChip’s Aether Core LLM accelerator on FPGA (cf. https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-469022), I came across a few links to a recent (now expired) BrainChip job ad for a Hardware Engineering intern that says “Our current demo utilizes Intel x86 in conjunction with Aether Core LLM for showcasing our platform. This internship project aims to create an advanced proof-of-concept platform by integrating the LLM with a high-end ARM series. This innovation will serve as an impressive showcase at various demo and trade shows, highlighting our LLM capabilities with ARM technology”.
Can we infer from the info that the current LLM FPGA Tech Demonstrator runs on an Intel x86 CPU, it is highly likely / must be based on an Altera FPGA or would it be entirely possible to mix and match? I have no idea…
Altera® FPGA Products - FPGA and SoC FPGA Devices and Solutions
Altera® FPGAs offer a wide variety of configurable embedded SRAM, high-speed transceivers, DSP blocks, high-speed I/Os, logic blocks, and routing.www.intel.com
View attachment 88201
View attachment 88200
The Roadmap shows that GenAI and Akida 2 are available in FPGA, and Akida 3 is "planned".
I think 6 nodes of FPGA was mentioned (in one of the electronics show videos ...?), so it's not the Rolls-Royce version, but would still pack a punch even in FPGA.
https://brainchip.com/brainchip-technology-roadmap/
Now there's an interseting thing I've just noticed:
Akida Gen AI has FP16 and FP 32;
Akida 3 has INT16, FP 16, and FP32.
So GenAI, which is adapted to process LLMs, does not need INT numbers. Or, to put it another way, LLMs are written in FP. GernAI is compatible with open source SSMs and LLMs.
We did use Xilinx many years ago for Brainchip Accelerator:
https://brainchip.com/brainchip-int...n-of-neuromorphic-computing-brainchip-120917/
SAN FRANCISCO, CA — (Marketwired) — 09/12/17 —
BrainChip Holdings Ltd. (ASX: BRN)
- Enables 16 channels of simultaneous video processing;
- Provides a low power, up to 6x speed boost to BrainChip Studio’s CPU-based Artificial Intelligence Software for Object Recognition;
- 7x more efficient than GPU-accelerated deep learning systems
...
The processing is done by six BrainChip Accelerator cores in a Xilinx Kintex Ultrascale field-programmable gate array (FPGA).
... but I don't think that that would be relevant, as FPGAs are pretty much COTS products. We would need an FPGA which had sufficient of the correct types of logic circuits.
Frangipani
Top 20
New repository set up yesterday on GitHub by NeuroSyd with the title:
Akida-Seizure.
Nothing added as yet but you'd suspect given what they do at the facility it would be our Akida they either been or will be playing with.
View attachment 83528
View attachment 83529
NeuroSyd Research Laboratory
NeuroSyd Research Laboratory at the Faculty of Engineering, The University of Sydney, is where emerging paradigms in computing, data science, neuro/bio-engineering and nanotechnology collide. NeuroSyd research is highly multidisciplinary and span over a wide range of fields from Device to Data,sites.google.com
View attachment 83530
Omid Kavehei
The University of Sydney - Cited by 6,831 - nanoelectronics - medical electronics - affective computing - learning machines - integrated circuit designscholar.google.com
View attachment 83531
Found a brand new University of Sydney PhD thesis, which is highly likely related to the GitHub “Akida Seizure” repository that one of the two thesis supervisors, Omid Kavehei, had set up under the GitHub name NeuroSyd just over two months ago (which seems to have disappeared since or has possibly been renamed?), spotted by @Fullmoonfever at the time.
Student Profile
University of Sydney PhD candidate Andre Zainal, who has a B.Eng. (Biomedical) & B.MedSci background, embarked on his Research PhD in Biomedical Engineering just a week ago, “researching AI-driven neuromorphic hardware for early seizure detection”.
The title of his thesis is “Optimizing Training Algorithms for Neuromorphic Hardware: Enhancing In-Memory Computing with FPGA and Akida Neural Processors for Epileptic Seizure Detection”.
Akida doesn’t have “in-memory-computing”, though?
Andre Zainal’s first PhD supervisor Omid Kavehei has been conducting epilepsy research for years:
Staff Profile
(…)
Similar threads
