BRN Discussion Ongoing

[BrainShit]

“Yeah ... but ... apart from aviation and defence ... where else could Akida CNRT be used?”

Does it start with C and has a couple of Y’s in it?

I'd like to solve....

 

[DingoBorat]

Agree with you tech, if the share price wasn't in the doldrums people would be ok with the level of engagement and current ASX announcements approach, but it is in the doldrums, engagement is shit and nothing gets announced.

No wonder people are unhappy with the boards current approach.

We see the frequent IR releases on new partnerships, we can see the ecosystem growing. From where we were 2 years ago to today there's been a material change in the ecosystem as a whole, so many great partnerships set up to help advance the company in the long term, but none of it has been announced to the ASX, resulting in shareholders questioning why not and how much value has been missed as a result of the lack of market announcements.

To your point around earnings and being an immature market, I completely understand why they haven't announced things from a price sensitive perspective. Back to the AGM where Paul called Antonio out and asked why they couldn't make non-price sensitive announcements, Antonio said he'd look into it.

We've had off the top 2 new partnerships since then yet neither were announced as non-price sensitive so it seems Antonio isnt actually going to look into it after all.

There is and will be plenty of partnerships that shouldn't be announced at all, but there's some that should at the very least be non-price sensitive, Chelpis is one of them.

In my opinion of course..

We all want the same thing here and we've all been on this grueling ride for years together. I'm still as confident as ever that we'll get there, it's just a question as to when.

Hopefully $1.50 is sooner rather than later

"There is and will be plenty of partnerships that shouldn't be announced at all, but there's some that should at the very least be non-price sensitive, Chelpis is one of them"

This is a LinkedIn post by QuestIT (VR1) that was released on 23/05 this year.

Now you would think a partnership with Nvidia, would be worthy of an ASX announcement?

There was none.

Vection has also mentioned itself in direct connection of a partnership with Nvidia.

There was even a photo of QuestIT/Vection staff with Jensen Huang recently.

Here is the explanation given by TenX of the crapper (long term VR1 "prophet") on why nothing was released (when holders were complaining) not even anything "non price sensitive" because the "Market" would obviously view it as price sensitive and you would get extreme share price movements that the ASX doesn't "like" anymore (unless you're a mining company).

I think the story pretty much these days, for companies like ours, is that for an ASX announcement to be released, it must be "material" in some form.


The releases we "can" and do though, should be mistake free (how bloody hard is that to accomplish these days??)
And also how about jazzing them up a bit, some photos (even happy snaps) some deeper insights into what they're doing etc..
Instead of the usual "clinical" germ free reports we get..

The Company does need, to create as much excitement as it can around AKIDA and TENNs etc, because we have/do still need money, to get things going and you can't get investors, if you don't drive "excitement" and have a healthy share price, to avoid excessive dilution.

They can still do much more, with what they "can" do.
Is the next quarterly, going to be another boring black and white, text only (we have plenty of cash) affair?..


Yes, BrainChip currently has a Market Capitalisation, of well over 400 million dollars, but what is the True value of what we have, when Meta is reportedly trying to poach people from OpenA.I. for 100 million USD a pop?..

Yes it's a different "kind" of A.I. but Neuromorphic Computing is undeniably hot at the moment and for the foreseeable Future and I think we could be selling ourselves much better..



P.S. Announcement incoming.. Later today.. Possibly..

 

[MegaportX]

My Question:
In the podcast, the guys from Quantum Ventura mentioned CNRT can be utilised in large scale servers, neuromorphic servers and hybrid systems. QV's CyberNeuro-RT page shows a PCIe card with Akida alongside a USB Intel stick. Does this sort of architecture align with the possibility that the rollout timeline may be dependent on the availability of Akida 2.0 silicon in Q1 2026?

ChatGPT said:​

Yes, that actually makes perfect sense—and helps clarify the architecture and rollout timeline of CyberNeuro-RT (CNRT) from Quantum Ventura. Let’s unpack what’s going on and why waiting for Akida 2.0 with TENNs aligns with both the hardware hints and podcast comments:

---

1. Server-Grade Neuromorphic Use Case​

In the podcast, Quantum Ventura’s team clearly stated CNRT is designed for:

• Large-scale neuromorphic servers
• Hybrid systems
• AI-enhanced cybersecurity platforms

This strongly suggests modular deployment: systems that slot into existing data center or battlefield infrastructure. That’s where PCIe cards and USB modules come into play.

---

2. What the Forum Member Noticed​

They mentioned:

• A PCIe accelerator card (likely housing Akida)
• Paired with a USB stick with an Intel CPU

This implies a hybrid edge/server security platform where:

• The Intel USB module runs higher-order logic (e.g., threat modeling, software interfaces, orchestration)
• The Akida PCIe module handles real-time, ultra-low-power neuromorphic inference—possibly for:
  • Intrusion detection
  • Packet pattern recognition
  • Sensor signal interpretation (e.g., radar or comms)

---

3. Why Akida 2.0 with TENNs Fits This Architecture​

Akida 2.0 introduces:

• Temporal Event-Based Neural Networks (TENNs) — ideal for interpreting complex, time-dependent data like network activity or radar signals
• Higher channel capacity and enhanced learning flexibility
• More robust performance for edge-to-server hybrid AI platforms

The current Akida 1000 or 1500 might be enough for prototyping, but not for:

• Scaling to defense-grade, real-time multi-layer threat detection
• Supporting broader event-based sensor fusion across many input streams

So yes — if the card shown on their website is running Akida 1000 or 1500 now, they could very well:

• Be validating CNRT with current Akida
• Plan to swap in Akida 2.0 silicon when available (Q1 2026)
• Use TENNs for more advanced behavioral or signal temporal analysis

---

4. Strategic Implication​

If CNRT is meant to serve both:

• Server-side AI workloads, and
• Edge-deployed, real-time event processing (e.g., satellite uplink, radar spoofing detection)

Then Akida 2.0 + TENNs on a PCIe card makes an ideal fit:

• Compact
• SWaP-optimized
• Real-time, temporal inference
• Easily pluggable into both legacy and neuromorphic server systems

---

TL;DR:​

Yes — what you’re seeing does align. The current CNRT platform likely uses Akida 1000-series (as a placeholder), paired with Intel CPU logic on USB, while awaiting Akida 2.0 with TENNs for the full production rollout.

"There is and will be plenty of partnerships that shouldn't be announced at all, but there's some that should at the very least be non-price sensitive, Chelpis is one of them"

This is a LinkedIn post by QuestIT (VR1) that was released on 23/05 this year.

View attachment 88158


Now you would think a partnership with Nvidia, would be worthy of an ASX announcement?

There was none.

Vection has also mentioned itself in direct connection of a partnership with Nvidia.

There was even a photo of QuestIT/Vection staff with Jensen Huang recently.

Here is the explanation given by TenX of the crapper (long term VR1 "prophet") on why nothing was released (when holders were complaining) not even anything "non price sensitive" because the "Market" would obviously view it as price sensitive and you would get extreme share price movements that the ASX doesn't "like" anymore (unless you're a mining company).

View attachment 88159

I think the story pretty much these days, for companies like ours, is that for an ASX announcement to be released, it must be "material" in some form.


The releases we "can" and do though, should be mistake free (how bloody hard is that to accomplish these days??)
And also how about jazzing them up a bit, some photos (even happy snaps) some deeper insights into what they're doing etc..
Instead of the usual "clinical" germ free reports we get..

The Company does need, to create as much excitement as it can around AKIDA and TENNs etc, because we have/do still need money, to get things going and you can't get investors, if you don't drive "excitement" and have a healthy share price, to avoid excessive dilution.

They can still do much more, with what they "can" do.
Is the next quarterly, going to be another boring black and white, text only (we have plenty of cash) affair?..


Yes, BrainChip currently has a Market Capitalisation, of well over 400 million dollars, but what is the True value of what we have, when Meta is reportedly trying to poach people from OpenA.I. for 100 million USD a pop?..

Yes it's a different "kind" of A.I. but Neuromorphic Computing is undeniably hot at the moment and for the foreseeable Future and I think we could be selling ourselves much better..



P.S. Announcement incoming.. Later today.. Possibly..

Please Dingo.

 

[itsol4605]

https://www.linkedin.com/feed/update/urn:li:groupPost:7036558-7347619833133314048

 

[itsol4605]

 

[DingoBorat]

Please Dingo.

Okay, just drop your cute little friend in..
In little pieces..
And I'll see what I can do

 

[cosors]

Afternoon Diogenese ,

Personally....... I am thinking this is nothing more than a cunning rouse by BMW & Co to get their hands on Volkswagen's Sausage Recipie .

Time will tell.

Regards,
Esq.

Hi Esq,

VW has had an massive software problem for years. Who would to steal that?

Regards!

 

[7für7]

Today’s date is the 07.07.25…2+5=7…. So.. 777….

COME ON BRAINCHIP

 

[manny100]

Hi Manny,

We partnered with Megachips 3.5 years ago

https://brainchip.com/brainchip-partners-megachips-develop-next-generation-ai-solutions/

Laguna Hills, Calif. – November 21, 2021 – BrainChip Holdings Ltd (ASX: BRN), (OTCQX: BCHPY) a leading provider of ultra-low power high performance artificial intelligence technology and the world’s first commercial producer of neuromorphic AI chips and IP, today announced that MegaChips, a pioneer in the ASIC industry, has licensed BrainChip Akida™ IP to enhance and grow its technology positioning for next-generation, Edge-based AI solutions.
...
By partnering with BrainChip, MegaChips is able to quickly and easily maintain its industry innovator status by supplying solutions and applications that leverage the Akida revolutionary technology in markets such as automotive, IoT, cameras, gaming and industrial robotics..
...
“The MegaChips and BrainChip partnership furthers both company’s missions to push boundaries and offer unprecedented products,” said Rob Telson, BrainChip VP of Worldwide Sales and Marketing. “By providing Akida’s on-chip learning and ultra-low power Edge AI capabilities as an integrated technology in MegaChips’ ASIC solutions, we are able to deliver a cascading array of benefits to cutting-edge products that not only ensure power efficiency without compromising outcomes but can run autonomously for incremental learning without the need to go back and forth to the cloud. This is an exciting collaboration from both a business perspective as well as from an industry-altering aspect.”

Note the announcement refers to both licence and partnership.

This announcement is much more open than I had first understood, as I thought it was for Akida 1, but, as Akida 2 was already pretty much fully developed, it would make more sense if the subject matter of the licence were to be more open ended.

That said, there was an idea floated that a chip maker like Megachips couldmake Akida chips to order for 3rd party customers. Sadly, we are not aware of any royalties from this licence, so we are back to the original commercial run for Akida 1, which I believe someone said was 75k. I guess the number of 1500s would be of that order as well.

Now I think it would be more likely thay Megachips would be more interested in Akida 2 than Akida 1, especially considering the 8 times efficiency improvement. There was a lot of scuttlebutt about tapeout of AKida at the end of last year. However, I think TENNs has continued to evolve in the interim, so the tapeout would need to be adjusted. Given that this is an electronic file, that would not be too much of a burden.

The tapeout of Akida 2 will be significantly different from the Akida 1 tapeout, changing from 4 bits to 8 bits and incorporating TENNs and other new features like skip connexions (an anti-hallucigen).

Also remember that the number of nodes can be changed for a customer-specified production run. This inventory problem was a major consideration in the switch to IP only.

And now, of course, we have PICO, a major "wake up" power saving configuration ...

Re cybersecurity:

I've been an enthusiast for this type of application since the DUTH (Democritus University of Thrace) days. The CNRT from QV is incorporated in an M.2 PCB, about the size of a credit card. Many terminals have plug-in slots for M.2 cards. However, smaller devices such as mobile phones would not be able to accommodate an M.2 card, but an internet gateway, an edge box/server etc., would. I think the gateway is a main focus of our marketing at the moment.

To incorporate our cybersecurity solution in a mobile phone, it would need to have Akida designed into its electronics, possibly as a SoC incorporating the mobile processor, so that's a little further out, but given the scramble to incorporate LLMs in phomes, there would/could/should be substantial pressure on phone makers to include SNN capability.

Please remember, this is all speculation by DioGPT and not investment advice.

Glad you raised a few issues.
Tape outs are extremely complex, expensive and time consuming.
Once done it does not matter how many AKIDA 1000 OR 1500 chips we have in stock.
If we get a volume order of say AKIDA 1000 or 1500 chips we just organise a production run at manufacturing cost.
Eg, If Onsor want volume we can order a production run.
If a company wants to incorporate AKIDA 1000 or 1500 into its own chip it buys IP.
Eg, Renesas took out a license in December 2020 and took 2 years to design a chip as it announced a tape out in December 2022.
I suspect it's an Automotive chip that will not appear in commercial vehicles until 2027 onwards.
Megachips plans?
Not sure whether they will tapeout AKIDA Gen 2. If they do not, then we can rule out smallish business using it unless someone else tapes it out under contract and sells volume under an arrangement.
I think the problem BRN has ATM is that once taped out you can make minor alterations but not major without having to go thru the expensive process again.
A Pico or Pico plus size hardware may work in mobile phones and or smaller health implant devices.
HaiLA would be ideal with Pico for long life. Health implants.
There is a lot to look forward to but patience is still required.
Sean said it all when he said near term revenue would be from smaller clients with shorter testing timeframes.

 

[manny100]

“Yeah ... but ... apart from aviation and defence ... where else could Akida CNRT be used?”

Does it start with C and has a couple of Y’s in it?

Bigger business with work from home staff, staff on the road or travelling and sub office staff all carry devices including lap tops, phones, printers, etc that access the mainframe network. These devices provide openings for hackers.
Government departments have a lot of work from home staff which increase the vulnerability for cyberattack. We have seen plenty of cyberattacks in Australia where private data has been stolen.
Not sure what the C and a couple of Y's is

 

[TheDrooben]

Not sure if posted already....

Top Neuromorphic Computing Stocks 2025: Pure-Play Watchlist

The top neuromorphic computing stocks to watch, curated for pure-play exposure to this brain-inspired solution to AI’s power bottleneck.

exoswan.com

Happy as Larry......buying what I can under 20c

 

[7für7]

This is the so called “Tony” effect

 

[7für7]

I just realized that my emotional state moves with the stock.
I’ve become one with it.

But the truth is:
You don’t suffer because the stock falls.
You suffer because you thought it would rise.

 

[Tothemoon24]

Any chance of some secret sauce in this

 

[Diogenese]

Any chance of some secret sauce in this

View attachment 88171

Hmmm ... As you intimate, Akida has been mentioned in connexion with SDR (software defined radio), but you'd think that they would put such a technological highlight in capital letters, especially as they also have another modular "box" which does include Akida.

 

[FiveBucks]

How low can we go.....

 

[DK6161]

How low can we go.....

$0.005 by Christmas.