BRN Discussion Ongoing

[manny100]

A pr

Antonio has previously implied that a single deal may turn us into a profit generating company but obviously it doesn't look like it will happen any time soon... (given Sean's target is 9Mill and give that figure was decided considering current engagements)...

By any chance, have you or anyone else managed to ask Antonio why he suggested during the AGM that they never specifically said it is the "US" that they are considering to redomicile? Even though the asx market sensitive announcement specifically said US..?

Why would you focus on the apparent 'gaffe'. Everyone makes them. Seems 'small picture' In relation to the business there is so much going on.
For example BRN has developed a prototype AKIDA model for TENNs/LLMs that should be ready in 12 months. That is huge especially when you consider Pico runs off TENNs.

 

[DK6161]

Agree, business wise it was very promising. All the 'he said/she said' carry on is just a smoke screen from the traction that we are gaining.
Many building early-stage companies have a bit of fire and brimestone at AGM's.
Best to keep or eyre on the business side of things.

Agreed. There shouldn't be any expectation at the very early stage for a disruptive tech company. Even 2034/35 seems a bit too early to expect any steady stream of revenue. I am happy to add another 5 of 6 years on top of that to give them a chance to double the current share price.
Akida everywhere
Not advice as always.

 

[Motty]

You are right, they should be informative and friendly, that’s why I informed you your question was bordering on stupidity ( informative) and I called you mate ( friendly ) Have an awesome night mate.

 

[Cardpro]

A pr

Why would you focus on the apparent 'gaffe'. Everyone makes them. Seems 'small picture' In relation to the business there is so much going on.
For example BRN has developed a prototype AKIDA model for TENNs/LLMs that should be ready in 12 months. That is huge especially when you consider Pico runs off TENNs.

My point was that there has always been some sort of expectation for the revenue ...... it's not a brand new idea that a business generates revenue... actually, the expectation was there even before Akida....... I even remember back in like 2018 or before people were wondering whether we r in Amazon Alexa lol...

 

[jrp173]

It's a forum mate,
I asked a question and someone was nice enough to answer myself,
I was at the AGM And I didn't quite get it so what,
Forums should be friendly and informative

Mate I'd ignore the person being obnoxious to you.

You'll soon see on this forum if you don't post what people like or if they disagree with you, then you are treated rudely or told to leave. Just ignore. Ask as many questions as you like, this is a public forum and I'm sure there are plenty who will answer questions if they can.

 

[jrp173]

A pr

Why would you focus on the apparent 'gaffe'. Everyone makes them. Seems 'small picture' In relation to the business there is so much going on.
For example BRN has developed a prototype AKIDA model for TENNs/LLMs that should be ready in 12 months. That is huge especially when you consider Pico runs off TENNs.

Manny to call this a "gaffe" is an understatement.

Should we not expect our Chairman, who was actually quoted in the announcement to know what he is talking about? Should we not expect accurate information?

This was an ASX released price sensitive announcement.

Personally, I like correct and factual information from our Execs and NEDS.

As already posted it speaks volumes that no-one from BrainChip through they should intervene to stop the train-wreck around the re-domicle questions...

 

[manny100]

Manny to call this a "gaffe" is an understatement.

Should we not expect our Chairman, who was actually quoted in the announcement to know what he is talking about? Should we not expect accurate information?

This was an ASX released price sensitive announcement.

Personally, I like correct and factual information from our Execs and NEDS.

As already posted it speaks volumes that no-one from BrainChip through they should intervene to stop the train-wreck around the re-domicle questions...

Focus on the gaffes and you miss many big picture positives such as:
Around the 25.30 mark of the 'Tech Talk'. Use cases of AKIDA 3. He says discussions with DOD and a "very high probability of getting a very positive outcome." This is in connection with every soldier wearing a headset with cameras interpreting what is around them, eg recognising hand signals from someone behind them. Evidently the DOD reps were speechless on hearing about this as communicating effectively in battle conditions has been a huge problem.
There are so many positives from the tech talk and AGM.

 

[Diogenese]

Lumi still offer the roadmap presentation if you cant wait for the release on the website.​

BrainChip Technology Roadmap Presentation​

View attachment 84062

It seems like the 'engine bay' is NOW ready for customers to play with, tweek and test out our wares in relation to their products.
Availability of model support and simpler set up will facilitate faster processes for the end user.
We seem to have been feeding them fish until now, we are offering them to learn how to fish going forward. Less 'hand holding'.


9 million bookings this year for the CEO as a KPI will be a good thing to see.
One might think employees are set achievable targets some might receive stretch targets I hope this is the former.
One would think 9 million is small change for this kind of tech but any money entering might be the ignition source we need.
I'm looking forward to seeing the changes to the web site which were also mentioned.
From cold dead hands I remain

Thanks FK,

The roadmap is chock full of groundbreaking advances. I felt that question time could have been better utilized by addressing the new opportunities these advances provide.

 

[equanimous]

" Before our next AGM our technical teams will achieve a ground braking milestone: the launch of the industry's first AI accelerator for State-Space models (SSM)"
Leader of the pack.

 

[Kingkong2015]

Since Akida is adding SSMs support. Brainchip will have more use cases and clients in areas listed in below table

Top 20 Most Popular State Space / Sequential Models (Including LLMs & Modern Hybrids)​

Sorted by real-world usage, research relevance, and adoption in devices:

Rank	Model	Type	Use Cases
	Large Language Models (LLMs)	Deep Learning	Chatbots, agents, assistants (GPT, Gemini, Claude)
	Kalman Filter	Linear-Gaussian	AR/VR head tracking, IMU fusion, finance, GPS
	Hidden Markov Models (HMMs)	Probabilistic	Speech recognition, gesture tracking, NLP tagging
	Extended Kalman Filter (EKF)	Nonlinear	Drones, autonomous navigation, smartwatches
	Particle Filter	Nonlinear, Sampling	AR/VR tracking, robotics, IoT localization
	Unscented Kalman Filter (UKF)	Sigma-point	Automotive ADAS, wearables, inertial sensors
	Structural Time Series Models (STSMs)	Interpretable	Power grids, IoT energy sensors, economic forecasting
	Dynamic Linear Models (DLMs)	Bayesian	Sales forecasting, wearables, health analytics
	Switching State Space Models (SSSMs)	Regime-switching	Fault detection in IoT, market phase modeling
	Deep State Space Models (e.g., RNN-SSM)	Deep Probabilistic	AR gesture prediction, speech synthesis, biosignal modeling

Emerging or Specialized Models (Ranks 11–20)​

Rank	Model	Type	Use Cases
1	Neural State Space Models (NSSMs)	Deep Neural ODEs	Physics-informed wearables, prosthetic control
1	Variational State Space Models (VSSMs)	Bayesian Deep Learning	EEG/fMRI analysis, AR/VR cognitive load estimation
1	LSTM State Space Hybrids	Seq2Seq + State	Smartwatch motion prediction, anomaly detection
1	Transformer-based SSMs	Attention with dynamics	Real-time AR/VR captioning, gesture translation
1	Nonlinear Autoregressive Exogenous Models (NARX)	Time Series	IoT weather sensors, predictive maintenance
1	Bayesian Filters (non-Kalman)	Probabilistic	Smart meters, context-aware wearables
1	Time-Varying Coefficient Models (TVCMs)	Dynamic Regression	IoT sensor drift correction
1	Reservoir Computing (Echo State Networks)	RNN-like	Low-power IoT edge devices, wearables
1	DeepAR (Amazon)	RNN Forecasting	Smart energy usage forecasting
	Spatio-Temporal State Space Models	Multivariate	Environmental sensor networks, smart cities

Summary​

• AR/VR and smartwear absolutely use SSMs, especially Kalman, UKF, and Particle Filters — just often embedded and not advertised.
• IoT devices use lightweight or approximate SSMs to balance performance and battery life.
• New deep learning hybrids (like Transformer + SSM or LLM + Sensor fusion) are emerging, especially with edge AI chips

SSMs are actually already used in these domains, but they’re often:

• Embedded inside sensor fusion or signal processing algorithms.
• Not marketed as “state space models” in commercial AR/VR/IoT products.
• Hidden behind layers of software (e.g., SLAM systems, Kalman-based IMU fusion).

For instance:

• AR headsets use Kalman/UKF to track user head/eye positions.
• Smartwatches use EKF/Particle Filters to smooth noisy biometric data.
• IoT sensors apply SSMs for anomaly detection or sensor calibration.

 

[Slade]

A new website will be unveiled shortly. " This will be followed by launch of a dedicated, developer focused companion site later this year"
"
This should expand our developer eco system.

I think they realize that Edge Impulse is no longer going to support them and they need their own developer platform.

 

[Fullmoonfever]

Appears in a recent preprint, that Basharat Ali has been running Akida, with others, as part of his / her work on cybersecurity.

Akida gives some pretty good results.

Kinda craps on NVIDiA.


Neuromorphic Quantum Adversarial Learning
(NQAL): A Bio-Inspired Paradigm for DNS over HTTPS Threat Detection
Basharat Ali
Nanjing University
Research Article
Keywords: Network Security, NQAL in Network Security, Network Protocols, Enhancing Network Security,
Enhancing DoH Protocol Security, Threats Detection in Encrypted Network, Cyber Attacks Detections
Posted Date: April 30th, 2025


Abstract Excerpt:

To overcome these complex issues, this work proposes a new architecture—Neuromorphic Quantum Adversarial Learning (NQAL)—a bio-inspired, zero-knowledge-supported detection
mechanism combining spiking neural networks (SNNs), quantum noise injection (QNI), and federated swarm intelligence to immunize, rather than detect, DoH-based attacks.

The method relies on a neuromorphic model employing Dynamic Spiking Graph Attention (DSGAT) and Spike-Timing-Dependent Plasticity (STDP) to encode encrypted traffic as dynamic spike trains to enable ultra-fast, energy-efficient inference on processors such as Intel Loihi and BrainChip Akida

Experiment set up Except:

Experiments were carried out on neuromorphic hardware platforms such as Intel Loihi 2 and BrainChip Akida that provide sub-millisecond latency with low-power event-driven processing characteristics.

Akida results related Excerpt:

Table 5: Hardware Deployment Metrics
Platform Accuracy Latency Power Throughput
GPU (NVIDIA V100) 89.2% 3.1 ms 45 W 1,200 QPS
TPUv4 91.5% 2.8 ms 32 W 1,500 QPS
Loihi 2 98.7% 0.9 ms 4 W 9,800 QPS
Akida 99.1% 0.7 ms 3 W 12,400 QPS

Outcome of Table 5:

Hardware Installation Metrics presents the excellent performance of our neuromorphic hardware solutions towards accomplishing peak performance for DoH security systems. When comparing Loihi 2 and Akaida to GPU platforms and TPU platforms depicts easily how changing towards neuromorphic chips invokes important boosts in terms of both accuracy and efficiency. Both the GPU (NVIDIA V100) and TPUv4 initiated with low performance at 89.2% and 91.5% accuracy, respectively, but when executed on Loihi 2, accuracy jumped dramatically to 98.7%, and a further improved 99.1% on Akida.

This increase in accuracy is accompanied by a drastic reduction in latency, from 3.1 ms for GPU to 0.7 ms for Akida, illustrating the real-time processing capability of the neuromorphic hardware.

Besides this, the power usage of the
Loihi 2 and Akida platforms—4 W and 3 W respectively—is a brilliant power efficiency against traditional GPU-based systems consuming 45 W. Throughput is also dramatically increased, with Akida being able to support 12,400 QPS, in strong contrast to the GPU’s 1,200 QPS.

Such results justify the single value of neuromorphic hardware as an approach for energy-efficient high-performance DoH anomaly detection and prove how
our new approach beats current systems and becomes the future standard for real-time system encrypted traffic protection[14].

Full paper HERE

 

[IloveLamp]

Semitech Semiconductor | Smart Grid & IoT Solutions

Discover Semitech Semiconductor's leading-edge solutions for smart grids and IoT, enabling robust and efficient communication technologies.

semitechsemi.com

#electronex #electronicdesign #powerlinecommunication | Semitech Semiconductor

What a great week at ElectroneX in Melbourne! We had many insightful conversations about the use of our PLC technology in a variety of industrial and transportation applications. We are looking forward to engaging with more Australian customers and partners. Thanks to the team for a job well...

www.linkedin.com

In the latest episode of our video series, Syntiant’s CEO Kurt Busch and Federal GM Ethan Wais share updates on our AI work for government, including a DIU success memo validating our tech in… | Syntiant Corp.

In the latest episode of our video series, Syntiant’s CEO Kurt Busch and Federal GM Ethan Wais share updates on our AI work for government, including a DIU success memo validating our tech in real-world military use. Highlights include: ✅ Warfighters can now run compact, efficient AI models...

www.linkedin.com

 

[IloveLamp]

💖Save the Date for 2025 Andes RISC-V CON Hsinchu- RePioneering the Future🎯 | Andes Technology

💖Save the Date for 2025 Andes RISC-V CON Hsinchu- RePioneering the Future🎯 📆Date: June 10 2025 🕛Time: 13:30-17:00 (GMT+8) 📍Venue: Sheraton Hsinchu Hotel, Taiwan 🌟Join us to discover the latest #RISCV trends, explore Andes’ innovations in #AI, #automotive, application processors, and...

www.linkedin.com

 

[IloveLamp]

💖Save the Date for 2025 Andes RISC-V CON Hsinchu- RePioneering the Future🎯 | Andes Technology

💖Save the Date for 2025 Andes RISC-V CON Hsinchu- RePioneering the Future🎯 📆Date: June 10 2025 🕛Time: 13:30-17:00 (GMT+8) 📍Venue: Sheraton Hsinchu Hotel, Taiwan 🌟Join us to discover the latest #RISCV trends, explore Andes’ innovations in #AI, #automotive, application processors, and...

www.linkedin.com

 

[TECH]

Good morning,

Look, some criticism is justified in my opinion during different stages of our development, from poor communication, poor management of
our funds, poor business planning, as in, not addressing traction issues early enough, while funds kept being pumped into sales teams who
clearly couldn't sell the idea, which I believe was a management issue myself, and possibly internal staff accountability issues etc.... but I
would suggest that many young start-ups battle similar growing pains.

I have been onboard for just shy of 10 years, and currently have a sense of calm surrounding this investment, are we in the best position
we have been in for years? yes, I believe so, a lot has been to do with education, not only of the semi-conductor industry or companies
actually, genuinely wanting to understand the real benefits of SNN technologies, but also our company as a whole, it's been a real learning
experience, which all groups appear to be coming out the other side of into clear air.

Peter's years of hard work and knock backs clearly show just how far advanced his research was, hence, we appear to be coming to that
inflection point, the bridge that was once too far, feels like that same bridge is close to completion as we as a company claim our share
of this new frontier, there's plenty to share around, our share of the pie will be huge moving forward I'm convinced of that, my opening
paragraph was fair I believe, but I, like many long termers have moved on, learnt from those early errors (growing pains) and can finally
see a horizon with a bright light emerging, our day is closing in, Brainchip will be successful, stay the journey with me and many others,
we all deserve to rise up as one!

Love our company.... Tech.

 

[FJ-215]

A short interview with ARM CEO Rene Haas. Worth a look as we are copying their business model.

 

[FJ-215]

Whoops, take 2.

Interview with ARM CEO Rene Haas.

https://www.bloomberg.com/news/videos/2025-05-08/arm-ceo-says-ai-is-not-a-bubble-video

 

[manny100]

Good morning,

Look, some criticism is justified in my opinion during different stages of our development, from poor communication, poor management of
our funds, poor business planning, as in, not addressing traction issues early enough, while funds kept being pumped into sales teams who
clearly couldn't sell the idea, which I believe was a management issue myself, and possibly internal staff accountability issues etc.... but I
would suggest that many young start-ups battle similar growing pains.

I have been onboard for just shy of 10 years, and currently have a sense of calm surrounding this investment, are we in the best position
we have been in for years? yes, I believe so, a lot has been to do with education, not only of the semi-conductor industry or companies
actually, genuinely wanting to understand the real benefits of SNN technologies, but also our company as a whole, it's been a real learning
experience, which all groups appear to be coming out the other side of into clear air.

Peter's years of hard work and knock backs clearly show just how far advanced his research was, hence, we appear to be coming to that
inflection point, the bridge that was once too far, feels like that same bridge is close to completion as we as a company claim our share
of this new frontier, there's plenty to share around, our share of the pie will be huge moving forward I'm convinced of that, my opening
paragraph was fair I believe, but I, like many long termers have moved on, learnt from those early errors (growing pains) and can finally
see a horizon with a bright light emerging, our day is closing in, Brainchip will be successful, stay the journey with me and many others,
we all deserve to rise up as one!

Love our company.... Tech.

Thanks Tech, my take below,
I think a 'prime' reason was that PVM's Neuromorphic invention was way ahead of its time. There were no products it could slip into.
No one had any use for it at all. So it was never going to sell.
Late 2021 Sean who had a mountain of experience was employed to commercialise the product. The ASX announcement that we were full on commercial was issued in Jan'22.
He had a 5 year plan approved by the BOD - 18 months shy of expiring.
An eco system has been built and it's a living system given its always growing. It's based on the premise that in the semi conductor industry 'no one stands alone' - if you do you perish and die. So well done.
I have no doubt that even the experienced Sean and the BOD were surprised at the time it has taken for industry to take up neuromorphic AI at the Edge for which we are the standout leader. The latest tech road map shows that we will remain the leader.
Humans change slowly unless it becomes necessary. The conflict in Europe and ME and loose war talk has defense build ups world wide.
That deemed necessary urgent DOD beef up is our opportunity.
Sean mentioned at the AGM that we are submitting proposals every month to the DOD. US AFRL, Navy transition to the Edge via Bascom Hunter.
Industry will start to full on follow the DOS and of course space program public useful tech.
IMO we get judged from here. The past has passed.
Times 'they are a changing' and we must take full advantage.

 

[White Horse]

Appears in a recent preprint, that Basharat Ali has been running Akida, with others, as part of his / her work on cybersecurity.

Akida gives some pretty good results.

Kinda craps on NVIDiA.


Neuromorphic Quantum Adversarial Learning
(NQAL): A Bio-Inspired Paradigm for DNS over HTTPS Threat Detection
Basharat Ali
Nanjing University
Research Article
Keywords: Network Security, NQAL in Network Security, Network Protocols, Enhancing Network Security,
Enhancing DoH Protocol Security, Threats Detection in Encrypted Network, Cyber Attacks Detections
Posted Date: April 30th, 2025


Abstract Excerpt:

To overcome these complex issues, this work proposes a new architecture—Neuromorphic Quantum Adversarial Learning (NQAL)—a bio-inspired, zero-knowledge-supported detection
mechanism combining spiking neural networks (SNNs), quantum noise injection (QNI), and federated swarm intelligence to immunize, rather than detect, DoH-based attacks.

The method relies on a neuromorphic model employing Dynamic Spiking Graph Attention (DSGAT) and Spike-Timing-Dependent Plasticity (STDP) to encode encrypted traffic as dynamic spike trains to enable ultra-fast, energy-efficient inference on processors such as Intel Loihi and BrainChip Akida

Experiment set up Except:

Experiments were carried out on neuromorphic hardware platforms such as Intel Loihi 2 and BrainChip Akida that provide sub-millisecond latency with low-power event-driven processing characteristics.

Akida results related Excerpt:

Table 5: Hardware Deployment Metrics
Platform Accuracy Latency Power Throughput
GPU (NVIDIA V100) 89.2% 3.1 ms 45 W 1,200 QPS
TPUv4 91.5% 2.8 ms 32 W 1,500 QPS
Loihi 2 98.7% 0.9 ms 4 W 9,800 QPS
Akida 99.1% 0.7 ms 3 W 12,400 QPS

Outcome of Table 5:

Hardware Installation Metrics presents the excellent performance of our neuromorphic hardware solutions towards accomplishing peak performance for DoH security systems. When comparing Loihi 2 and Akaida to GPU platforms and TPU platforms depicts easily how changing towards neuromorphic chips invokes important boosts in terms of both accuracy and efficiency. Both the GPU (NVIDIA V100) and TPUv4 initiated with low performance at 89.2% and 91.5% accuracy, respectively, but when executed on Loihi 2, accuracy jumped dramatically to 98.7%, and a further improved 99.1% on Akida.

This increase in accuracy is accompanied by a drastic reduction in latency, from 3.1 ms for GPU to 0.7 ms for Akida, illustrating the real-time processing capability of the neuromorphic hardware.

Besides this, the power usage of the
Loihi 2 and Akida platforms—4 W and 3 W respectively—is a brilliant power efficiency against traditional GPU-based systems consuming 45 W. Throughput is also dramatically increased, with Akida being able to support 12,400 QPS, in strong contrast to the GPU’s 1,200 QPS.

Such results justify the single value of neuromorphic hardware as an approach for energy-efficient high-performance DoH anomaly detection and prove how
our new approach beats current systems and becomes the future standard for real-time system encrypted traffic protection[14].

Full paper HERE

BrainChip is doing business in Korea, Japan and Taiwan. We have been working with international companies like Renesas. It is hard to imagine that China hasn’t had a good look at Akida. Being an Australian company we are not restricted in dealing directly with China by US sanctions. If we have stepped back from dealing with China to please the US then I hope it proves worth it.

Hi Slade, remember this from 2022.
https://www.ex3.simula.no/resources

Posted by FMF.
https://thestockexchange.com.au/thr...rver-kunpeng-920-processor-using-akida.29899/

Well, if anybody was doubting interest from China.
I'd say the answer is YES.!!!