BRN Discussion Ongoing

[manny100]

Its been almost 18 months since TATA announced the partnership so we must be closing in on deal. They did say they intend to use AKIDA.

Tata Elxsi - BrainChip and Tata Elxsi Partner to Provide Intelligent, Ultralow-Power Solutions

Tata Elxsi

www.tataelxsi.com

 

[TECH]

Gidday back in Australia,
I must first apologise for misleading posters whom liked my post regarding Boardroom Pty Ltd.

I was of the understanding that we could approach the registry and on providing our indentity could get access to the Top 20 holders within our company, whether there has been a change in policy I'm not sure, but I find it rather absurd.

The EETimes podcast was excellent, probably beyond many of us on a technical level, plenty of positive chatter, but once again, it's a real journey, I'm a little more conservative on future stock price guess's..over $1 by end 2025, pushing the $2 mark by mid 2026 would satisfy my business mind.

But plenty of twists and turns yet to play out, and lets please cease with all the 2nd strike AGM chatter again like this time last year...deal with that issue around AGM time in late May, early June...thanks.

Tech.

 

[manny100]

An IP deal can’t be too far away with Tata

View attachment 77417

View attachment 77416

Multimodal AI and Neuromorphic AI: Detection, Diagnosis, Prognosis - Tata Elxsi

Discover how Multimodal and Neuromorphic AI lead healthcare from reactive to proactive. This article delves into Responsible AI and its transformative impact.

ai.tataelxsi.com

Multimodal AI and Neuromorphic Technology – A new era in Preventive Healthcare​

Tata said they were driving AKIDA into Health and Industrial.

 

[JB49]

Careers at Northrop Grumman

Careers at Northrop Grumman

www.northropgrumman.com

Northrop Grumman in the US is looking for someone with an interest in neuromorphic computing. But unfortunately their focus is on analog neuromorphic.

Then they have 2 scientist roles for neuromorphic computing in the cyber security space.

Careers at Northrop Grumman

Careers at Northrop Grumman

www.northropgrumman.com

Careers at Northrop Grumman

Careers at Northrop Grumman

www.northropgrumman.com

 

[Braintonic]

Correct me if I am wrong but are the latest announcements in relation to Akida 1000? And the company said it can take a couple of years of support to build a system that suits the client.
Secondly they listened to their customers and then built Akida 1500 and 2 and Tenns and Pico.
I am buoyant that will start to see the rewards of Akida 1000 this year but with years to come, more engagements more products and more profits we will definitely start seeing more rewards.
Just my thoughts.

 

[manny100]

Gidday back in Australia,
I must first apologise for misleading posters whom liked my post regarding Boardroom Pty Ltd.

I was of the understanding that we could approach the registry and on providing our indentity could get access to the Top 20 holders within our company, whether there has been a change in policy I'm not sure, but I find it rather absurd.

The EETimes podcast was excellent, probably beyond many of us on a technical level, plenty of positive chatter, but once again, it's a real journey, I'm a little more conservative on future stock price guess's..over $1 by end 2025, pushing the $2 mark by mid 2026 would satisfy my business mind.

But plenty of twists and turns yet to play out, and lets please cease with all the 2nd strike AGM chatter again like this time last year...deal with that issue around AGM time in late May, early June...thanks.

Tech.

Welcome back TECH, if there is no more positive news between now and the end of the year........................ the SP will be cactus. But if we get a steady flow of news as expected sentiment can change very quickly.
We saw a steady flow of positive news move the SP from a September low of 15.5 cents to a January high of 45 cents. That was sentiment driven. Market calms and back to 28 cents on Friday. If we have the year that Sean promised and we had that number of engagements that Tony V mentioned at the last AGM plus engagements since we might see a steady flow of news for the rest of the year. SP - depends on how good and how regular the news is.
We did not know about interest in the AKIDA 1500 until the mention of Bascom Hunter in the last 4c. buying $100k worth of chips which gets them a few. The chips are for full scale evaluation of commercial products which is getting serious given the volume they bought. Its not just factory trials it looks more like field trials which is getting close to full on.
AKIDA makes BH Tech look even better.
AKIDA makes QV cybersecurity 'the only game in town'. I expect the Lockheed-Martin will be marketing it to its huge Defense client base.
AKIDA makes Frontgrade the best in Space.
TATA have publicly stated that they will drive AKIDA into Health and Industrial because they know we will make them look good.
The new EDGE box ecosystem will make our boxes very versatile.
There is likely going to be some out of left field news this year as well.
and then there is the US AFRL contract.
and Renesas taped out chips - destination?
and we have not yet heard a word about moving GEN2, TENNs which have been available to developers for a while.
I am bullish on BRN and not shy about it. If i wasn't i would sell.
Sean and Tony V have been taking huge chunks of their salary as RSU's. I guess like me they are bullish.

 

[Diogenese]

Correct me if I am wrong but are the latest announcements in relation to Akida 1000? And the company said it can take a couple of years of support to build a system that suits the client.
Secondly they listened to their customers and then built Akida 1500 and 2 and Tenns and Pico.
I am buoyant that will start to see the rewards of Akida 1000 this year but with years to come, more engagements more products and more profits we will definitely start seeing more rewards.
Just my thoughts.

Yes. The forecast 16-bit version for many more algorithms and the "thinking" neurons are exciting prospects ... and the expected completion of Akida 2 architecture is something we've been hoping for for a while.

 

[manny100]

For any new readers who may wish to get some quick references to the BRN/QV/Lockheed-Martin only game in town cybersecurity see links below:

BrainChip Provides Low-Power Neuromorphic Processing for Quantum Ventura’s Cyberthreat Intelligence Tool

BrainChip Provides Low-Power Neuromorphic Processing for Quantum Ventura’s Cyberthreat Intelligence Tool Laguna Hills, Calif. – January 16, 2025 – BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, ADR: BCHPY), the world’s first commercial producer of ultra-low power, fully digital, event-based...

brainchip.com

BrainChip Powers Cyberthreat Tool

The integration of BrainChip's ultra-low power neuromorphic AI into Quantum Ventura's CyberNeuro-RT technology is poised to revolutionize cybersecurity protection for network access devices. By leveraging the Akida processor, this innovative solution enables on-chip learning and threat detection...

quantumzeitgeist.com

https://brainchip.com/wp-content/uploads/2025/01/BrainChip_White-Paper-A-Game-Changer-in-AI-Computing-for-Cybersecurity_v3.pdf?utm_source=website&utm_medium=DG+Landing+Page&utm_campaign=AI+Computing+for+Cybersecurity

 

[Diogenese]

Welcome back TECH, if there is no more positive news between now and the end of the year........................ the SP will be cactus. But if we get a steady flow of news as expected sentiment can change very quickly.
We saw a steady flow of positive news move the SP from a September low of 15.5 cents to a January high of 45 cents. That was sentiment driven. Market calms and back to 28 cents on Friday. If we have the year that Sean promised and we had that number of engagements that Tony V mentioned at the last AGM plus engagements since we might see a steady flow of news for the rest of the year. SP - depends on how good and how regular the news is.
We did not know about interest in the AKIDA 1500 until the mention of Bascom Hunter in the last 4c. buying $100k worth of chips which gets them a few. The chips are for full scale evaluation of commercial products which is getting serious given the volume they bought. Its not just factory trials it looks more like field trials which is getting close to full on.
AKIDA makes BH Tech look even better.
AKIDA makes QV cybersecurity 'the only game in town'. I expect the Lockheed-Martin will be marketing it to its huge Defense client base.
AKIDA makes Frontgrade the best in Space.
TATA have publicly stated that they will drive AKIDA into Health and Industrial because they know we will make them look good.
The new EDGE box ecosystem will make our boxes very versatile.
There is likely going to be some out of left field news this year as well.
and then there is the US AFRL contract.
and Renesas taped out chips - destination?
and we have not yet heard a word about moving GEN2, TENNs which have been available to developers for a while.
I am bullish on BRN and not shy about it. If i wasn't i would sell.
Sean and Tony V have been taking huge chunks of their salary as RSU's. I guess like me they are bullish.

Hi manny,

The EE Times podcast has foreshadowed some exciting developments, and I expect we will see some patents filed for them soonish.

In particular the "thinking" neuron will add another dimension to Akida/TENNs capabilities.

Then there's the 16-bit weights/activations which reportedly will facilitate new algorithms.

... I've got a nose-bleed thinking about it.

... and then the architecture of Akida 2 is being finalized, quite a while after the false dawn of tape-out a while ago. That indicates to me that there may have been some additional performance enhancements in the interim - maybe radhard, or ...?

Given the number of space and defence-related implementations, it may be that it has been decided to build in radhardness for Akida2???

 

[Diogenese]

Hi manny,

The EE Times podcast has foreshadowed some exciting developments, and I expect we will see some patents filed for them soonish.

In particular the "thinking" neuron will add another dimension to Akida/TENNs capabilities.

Then there's the 16-bit weights/activations which reportedly will facilitate new algorithms ... I've got a nose-bleed thinking about it.

... and then the architecture of Akida 2 is being finalized, quite a while after the false dawn of tape-out a while ago. That indicates to me that there may have been some additional performance enhancements in the interim - maybe radhard, or ...?

Given the number of space and defence-related implementations, it may be that it has been decided to build in radhardness for Akida2???

The helicopters!!! I shouldn't have mentioned radhard!

 

[manny100]

This is what i mean when i say AKIDA 1000 makes the Bascom Hunter 3UVPX SNAP CARD look better than good.
"Unlike traditional machine learning accelerators such as GPUs or TPUs, neuromorphic processors are designed to mimic the biological efficiency of the human brain, allowing the SNAP Card to run multiple ML models in parallel at exceptionally low power — just 1W per model — without sacrificing speed or accuracy while the FPGA enhances input/output operations and performs any desired signal processing tasks."

AKIDA allows BH to do it all simultaneously in real time and ultra low power..
1. automatic target recognition,
2. Real time detection and identification of threats.
3. Detection of FISINT (Foreign Instrumentation Signature Intelligence) and/or hacking across the airframe’s
4. Perform communications analysis to include speech-to-text and foreign language translation of intercepted communications
5. Fuse multiple infrared cameras (e.g., SWIR, MWIR, LWIR) to provide a combined infrared operating picture on the ground Multi-user, simultaneous modulation/demodulation
Easy to see why the DOD is transitioning to Neuromorphic AI and BH intends to be a Defence leader.
Its amazing how AKIDA1000 can do all that simultaneously in real time. Looks like a winner to me.......... and GEN2 plus TENNs will be an improvement again.
GEN 3 when it arrives with 16 bit plus whatever extra surprises will take us to an even higher level.
AKIDA makes its partners better.
See BH's 3U VPX product review link below.

https://bascomhunter.com/wp-content/uploads/2024/11/3U-VPX-SNAP-Card-Brochure-11252024.pdf

 

[FiveBucks]

The helicopters!!! I shouldn't have mentioned radhard!

The share price hitting $2 again will make me radhard!

 

[Diogenese]

This is what i mean when i say AKIDA 1000 makes the Bascom Hunter 3UVPX SNAP CARD look better than good.
"Unlike traditional machine learning accelerators such as GPUs or TPUs, neuromorphic processors are designed to mimic the biological efficiency of the human brain, allowing the SNAP Card to run multiple ML models in parallel at exceptionally low power — just 1W per model — without sacrificing speed or accuracy while the FPGA enhances input/output operations and performs any desired signal processing tasks."

AKIDA allows BH to do it all simultaneously in real time and ultra low power..
1. automatic target recognition,
2. Real time detection and identification of threats.
3. Detection of FISINT (Foreign Instrumentation Signature Intelligence) and/or hacking across the airframe’s
4. Perform communications analysis to include speech-to-text and foreign language translation of intercepted communications
5. Fuse multiple infrared cameras (e.g., SWIR, MWIR, LWIR) to provide a combined infrared operating picture on the ground Multi-user, simultaneous modulation/demodulation
Easy to see why the DOD is transitioning to Neuromorphic AI and BH intends to be a Defence leader.
Its amazing how AKIDA1000 can do all that simultaneously in real time. Looks like a winner to me.......... and GEN2 plus TENNs will be an improvement again.
GEN 3 when it arrives with 16 bit plus whatever extra surprises will take us to an even higher level.
AKIDA makes its partners better.
See BH's 3U VPX product review link below.

https://bascomhunter.com/wp-content/uploads/2024/11/3U-VPX-SNAP-Card-Brochure-11252024.pdf

Hi manny,

I think we're singing off the same hymn sheet. The BH deal might be for testing. Any warfighter processor must be radhard ... proof against airsickness (oops! - nearly said it again).

 

[manny100]

Hi manny,

The EE Times podcast has foreshadowed some exciting developments, and I expect we will see some patents filed for them soonish.

In particular the "thinking" neuron will add another dimension to Akida/TENNs capabilities.

Then there's the 16-bit weights/activations which reportedly will facilitate new algorithms ... I've got a nose-bleed thinking about it.

... and then the architecture of Akida 2 is being finalized, quite a while after the false dawn of tape-out a while ago. That indicates to me that there may have been some additional performance enhancements in the interim - maybe radhard, or ...?

Given the number of space and defence-related implementations, it may be that it has been decided to build in radhardness for Akida2???

Its more than exciting IMO. Even AKIDA 1000 with BH as per my previous post is a mind boggling leap from what traditional systems can do.
Agree 16 bit and 'thinking neurons' is huge and will keep us well in the lead with GEN3.
No one is even close IMO in relation for our use purposes (eg defense) to AKIDA 1000 as it is. We are targeting huge markets.
I guess with Gen 2 not in silicon yet and no apparent deals we are able to keep making enhancements. Likely from suggestions during engagements or developers playing with it?
Maybe one of the reasons for no GEN 2 Tape out?.
Developers have had access to GEN 2 plus TENNs for nearly 2 years so we may see a deal or 2 or at least significant interest news this year?

 

[manny100]

Hi manny,

I think we're singing off the same hymn sheet. The BH deal might be for testing. Any warfighter processor must be radhard ... proof against airsickness (oops! - nearly said it again).

Yes Dio, agree, i think its starting to 'gel' in the market that AKIDA is the COG/chip that makes others wheels turn.
We make our partners look unbeatable.
TL when asked in an interview at CES 25 what will AKIDA be in within 5 years said with a smile "everything" - no doubt a little tongue in cheek.
I admit to being unashamedly bullish.

 

[Diogenese]

Yes Dio, agree, i think its starting to 'gel' in the market that AKIDA is the COG/chip that makes others wheels turn.
We make our partners look unbeatable.
TL when asked in an interview at CES 25 what will AKIDA be in within 5 years said with a smile "everything" - no doubt a little tongue in cheek.
I admit to being unashamedly bullish.

Re TL's dental gymnastics, many years ago, in another (tautological) time and place, I foresaw Akida as being as ubiquitous as the 555 timer chip.

 

[itsol4605]

Great podcast at EE Times together with Brainchip

https://www.eetimes.com/podcasts/brainchips-ip-for-targeting-ai-applications-at-the-edge/

 

[Diogenese]

An IP deal can’t be too far away with Tata

View attachment 77417

View attachment 77416

As Slim Dusty said:

"We love to have a drink with Arpan,
'cause Arpan's our Pal.
We drink in moderation,
and we never ever, ever
have more than TENNs,
but we'll have one for the node.
We like to network neurally,
in parallel and plurally.
We love to have a drink with Arpan,
'cause Arpan's our Pal."

 

[Diogenese]

Pinched this from FF over on the crapper.

https://www.eetimes.com/podcasts/brainchips-ip-for-targeting-ai-applications-at-the-edge/

A video and transcript. I did the read.
Exract: Note my bolded area:
" Jon Tapson (JT): At BrainChip, we are currently finishing off our Akita 2.0 architecture. And this is an expansion on all of BrainChip’s prior work. It’s essentially a neuromorphic chip, it’s event-based, and it makes use of the intrinsic sparsity in signals to achieve high levels of efficiency.But we have the same problem that any company has that does actual… there’s an enormous difference between research silicon, if I can put it that way, and commercial silicon. And the problem with commercial silicon is you’ve got to fit into the ecosystem of what everybody else is doing. The history of startup companies is littered with companies that had fantastic technology, but couldn’t progress across the interface to regular technology. The real problem is if you build a chip, the chip by itself is not a solution to anything. It’s got to go on a board, it’s got to have a software ecosystem around it. And the easier that is for customers and engineers to access, the better everything’s going to go."
From my bold above its easy to see why AKIDA 1000 and 1500 are now gaining traction. AKIDA is no longer in research and actually commercial and we fit into the eco system of what everyone else is doing.
Its taken time but we are seeing the proof unfold before our eyes right now.
The AKIDA1000 fit with QV gives us the 'only game in town" cybersecurity with a Defence name in Lockheed- Martin to sell it.
AKIDA 1000 and more recently the 1500 gets the wheels of Bascom Hunter moving and they aim to be a leader in Defence.
AKIDA1000 gives its partner Frontgrade the 'frontrunning' in Space.
I expect to see more of this as engagements that have been on for quite some time come to fruition.
We make our eco system partners better.

Tony Lewis makes the point that "events" and "spikes" are different, spikes being singal-valued occurrences, while events each have an individual magnitude value.

This sort of aligns with my thoughts on why Jason Eschragian stated that he did not think Akida was "neuromorphic" in that analog SNN developers think spikes as single valued and accumulate in neurons until they reach the firing threshold, even thought Edgar Adrian's 1920s paper, the inspiration for Simon Thorpe's N-of-M coding, showed there is more to it than that. A lot of the early analog SNNs used rate coding, the rate at which spikes repeated or oscillated. Thorpe utilized the relationship between spike amplitude and timing.

 

[charles2]

The share price hitting $2 again will make me radhard!

If you desire premier 'rad-hardness' entice NVDA (or near equivalent) to take a 10% position in the company.

That should maintain this board in a state of hardness for many days and nights.

A lifetime perhaps.