We Taught Machines to See and Hear. Smell Was Always the Hard One.
Jason Dussault
Chief Executive Officer, Co-Founder
.jpg)
This Post is disseminated on behalf of Intellistake Technologies Corp.
A trained detection dog can walk the length of a shipping container and tell its handler that something inside it is wrong. It does not open the container. It does not send a sample to a laboratory. It does not wait three days for a result. It just knows, and it knows in seconds.
We have been trying to build a machine that does that same job for decades.Cameras got extraordinarily good. Microphones got extraordinarily good. The third one stayed stubborn.
On August 4, 2026, we announced that Intellistake had entered into a definitive agreement to acquire a business that has spent years working on exactly that stubborn problem.1
The two senses already solved
Think about what a camera actually does. It captures light. Light is one signal, it behaves predictably, and we have known how to measure it precisely for well over a century.
A microphone captures pressure waves. Same story. One signal, well understood, measurable to a very fine degree.
Once you have a clean signal, the rest is engineering and patience. And we have had a lot of both.
Machine vision now runs your phone camera, your car's lane assist, and a large amount of the world's industrial inspection. Speech recognition sits in every home that owns a smart speaker.So the natural question is why the third sense did not follow simply.
Why reading the air is a different kind of problem
Here is the part I find genuinely interesting, and I say that as someone who came to this from the software side rather than from chemistry.
Scent is not one signal. It is a crowd.
A single odour is typically made up of anywhere from dozens to thousands of different molecule types, drawn from a pool of more than 400,000 known odorous and odourless molecules.2
The air in an ordinary room carries hundreds of volatile organic compounds at once, from cleaning products, furniture, cooking, people, the building itself.3 The United States Environmental Protection Agency puts the measurement problem about as plainly as a government agency ever puts anything:
"all available measurement methods are selective in what they can measure and quantify accurately, and none are capable of measuring all VOCs that are present."4
So the task is not "measure the thing." The task is "find one specific pattern hiding inside an enormous, shifting crowd of other patterns, and do it while the temperature changes, the humidity changes, and the background composition changes by the minute."
This is a pattern recognition problem. Which is precisely why this field started moving when machine learning got good, and not before.
Biology, for what it is worth, figured this out a long time ago. Richard Axel and Linda Buck won the 2004 Nobel Prize in Physiology or Medicine for working out how the olfactory system is organised.5 Human beings carry somewhere in the region of 390 odorant receptor genes that appear to be functional.6 No single one of them identifies anything on its own. The identification happens afterwards, in the interpretation. I think that is the whole lesson, honestly, and it took the technology industry an embarrassingly long time to arrive at the same conclusion.
The peer-reviewed literature on artificial olfaction has been candid about the difficulty for years. Sensor stability and drift over time have been described as
"a constant concern in the community and a barrier to commercial adoption."7
That is a polite way of saying that a lot of devices work beautifully in a laboratory and then behave differently three months later in a warehouse in July.
Meanwhile, dogs kept doing the job. A 2016 review in Frontiers in Veterinary Science recorded canine detection down to parts per trillion, and noted at the time that this was around three orders of magnitude better than the instruments then available.8
Dogs also have limits nobody likes to discuss. Research published in 2021 found that unacclimatized explosive detection dogs perform optimally for only about fifteen to twenty minutes before performance fades quickly, with heat and humidity making it worse.9 A dog is an outstanding sensor with a short shift, a handler, a salary, a retirement date, and a bad day now and then.
There has been real movement recently. In 2023 a team published a "principal odor map" in Science, a machine learning model that predicted how molecules would be perceived and outperformed the median human panellist on a majority of the molecules tested.10 In 2024, researchers at the University of Pittsburgh published work on a machine learning approach to detecting and distinguishing fentanyl at very low concentrations.11 Early-stage laboratory research, not a deployed product, and it deserves to be described that way. But the direction is not ambiguous any more.
What I am deliberately not going to tell you
I want to be straightforward about something, because I would rather flag it than have you notice the omission and wonder.
I am not going to describe how the NanoAi detection technology actually works.
Our announcement describes the platform in general terms, and that is as far as I am going to take it here. Not the specifics of the materials. Not the internal design. Not the signal handling, not the model architecture, not the thresholds. Nothing that would let a competitor shortcut years of work, and nothing that would help somebody figure out how to defeat it.
Technology in this category exists to find explosives, chemical agents, biological threats and illicit drugs. If you publish enough about how a detection system reaches its conclusion, you have published a partial manual for evading it. Every serious person in this sector understands that trade-off, and it is why the detailed material stays in due diligence rooms and in conversations that do not happen in public. Perhaps that reads as evasive. I would rather be thought evasive than be genuinely useful to the wrong reader.
What I can talk about is what the capability does, and why anyone should care.
Why anyone should care
Start with the money, because for a cold reader that is usually the fastest way to understand whether a sector is real.
World military expenditure reached US$2,887 billion in 2025, a rise of 2.9 percent in real terms and the highest level SIPRI has ever recorded.12
In June 2025, NATO allies committed to invest 5 percent of GDP annually on core defence and defence-related spending by 2035, split between at least 3.5 percent for core defence requirements and up to 1.5 percent for infrastructure, resilience, innovation and the defence industrial base.13
In the United States, the FY2026 national defense topline came in at approximately US$1,011.9 billion,14 and the FY2027 request has been put at US$1.5 trillion.15
Third-party research puts artificial intelligence and analytics in defence at approximately US$13.95 billion in 2026, rising to approximately US$23.5 billion by 2030.16 A separate forecast puts the broader military AI market at approximately US$41.6 billion by 2035.17 Global spending on counter-drone systems has been estimated at US$12.6 billion in 2026, growing to US$24.1 billion by 2030, although I would note that this particular forecast comes from a trade publication rather than an independent research house and sits above some other estimates.18
Now the part the spending charts do not capture.
In September 2025, drone sightings shut Copenhagen Airport for around four hours, and Aalborg closed days later, with further sightings at several Danish sites including an air base.19
In November 2025, drones over Brussels Airport suspended all air traffic, with additional sightings reported over Belgian military installations.20 Press reporting in 2026 on research published by the International Institute for Strategic Studies described drones operating with what the institute called substantial impunity across European airspace, including over bases connected to NATO's nuclear mission.21
Attribution in these cases remains suspected rather than established, and I want to be careful to say so.
Then there is the drug problem, which is a different shape entirely. The CDC reported 69,973 provisional overdose deaths in the United States in 2025. That is a decline of almost 14 percent, the third consecutive annual fall, which is genuinely good news and does not remove the problem.22 The Department of Homeland Security announced in May 2026 that Customs and Border Protection had seized more than 100 million lethal doses of fentanyl along the southwest border in fiscal 2026 to that point.23
Consider what "seized" implies. Somebody had to find it. At a port of entry handling enormous volume, with limited screening time per vehicle, per container, per parcel. Screening capability is the constraint on that entire system, and everyone working in it knows it.
Every defence and security system deployed at scale today is built around sight and sound.
Cameras, radar, satellites, acoustic arrays, drones watching drones. Billions invested in the ability to see and hear a threat from a distance.
But an explosive leaves traces in the air. So does fentanyl. So do nerve agents, industrial chemicals and biological material. Those threats are invisible to a camera, silent to a microphone and absent from a radar return. That is the gap. It sits inside a market worth tens of billions of dollars a year, and it is the one part of the sensing stack that has not been properly closed.
What we announced
On August 4, 2026, Intellistake announced a definitive agreement, dated August 3, 2026, to acquire NanoAi Technologies. NanoAi is a nanotechnology and artificial intelligence company that has developed proprietary standoff detection devices capable of identifying multiple specific threats. Its stated applications run across defence, healthcare, aerospace, energy and critical infrastructure.1
The terms, briefly, Intellistake would acquire 100 percent of the outstanding securities of NanoAi in exchange for approximately C$17 million in Intellistake common shares, priced at C$0.50 per share, being 34,106,412 shares. Those shares are to be issued subject to the attainment of certain performance-based vesting milestones related to future contracts and revenues generated by NanoAi, and will be subject to escrow and contractual trading restrictions.
The transaction is arm's length and no long-term debt is being assumed. Subject to no objection from the Canadian Securities Exchange, Intellistake is to provide NanoAi with a bridge loan of US$1,300,000 at 10 percent per annum, advanced in instalments against work plan milestones. Completion remains subject to customary conditions including satisfactory due diligence, verification of title to NanoAi's intellectual property, completion of an audit of NanoAi's financial statements, Intellistake maintaining a minimum cash balance of $2 million, and no objection from the Canadian Securities Exchange. Closing is targeted within 60 days.1
I have put the conditions in deliberately. This is a signed definitive agreement, not a closed transaction, and the difference is not a technicality.
On NanoAi itself, staying strictly to what has been disclosed. The company has completed more than 60,000 validation tests for infection detection, with results delivered in approximately 30 seconds. I want to be precise about that number, because it is a specific figure attached to a specific application, and it should not be read as validation of every use case named above.
Its platform features proprietary standoff detection, identifying multiple distinct threats simultaneously in an ultralight form factor with no physical contact required. The platform extends beyond detection hardware into backend machine learning, geolocation and triangulation, and real-time data processing, supported by proven manufacturing scale. The company is headquartered in Dallas, Texas.1
In addition, NanoAi was founded by Craig Micklich, who served twelve years as a United States Navy SEAL, is a service-connected disabled veteran and a lifetime member of the UDT/SEAL community, and afterwards held senior roles in institutional finance including Managing Director at Deutsche Bank and Senior Vice President at Morgan Stanley. He has led the underlying sensing technology and business since 2018, and he is a co-author on peer-reviewed research underlying the core technology.1
My first impression of Craig, which has not changed, was that he is one of the very few people who has operated in the environments this equipment is built for and can also hold a serious conversation about capital structure. That combination is rarer than it sounds. On closing, Craig and a second NanoAi nominee would join the Intellistake board.1
What we bring, and why it is the software half
Here is where I get back to my opening argument.
For decades the assumption in this field was that the constraint was chemistry. Build a more sensitive material and the problem dissolves. That turned out to be wrong, or at least incomplete.
The harder constraint was interpretation. The signal was collectable. Reading it reliably, quickly, and in messy real-world conditions was the thing nobody had cracked.
That is the same structural problem we already work on every day, in a completely different domain. Enterprises are drowning in data they already own. Regulatory filings, market feeds, operational logs. The information exists.
The ability to turn it into a clear, verifiable answer fast enough to act on does not. That gap is what IntelliScope is being built to close.
Different domain, near identical structure. Raw signal is abundant and close to worthless. Trustworthy interpretation is scarce and very valuable.
So what do we actually bring. We bring the software half. We already run enterprise AI deployments, retrieval-augmented generation architecture and AI-driven data pipelines, and the intention is to point that capability at NanoAi's platform so it can handle many sensors and many devices at once rather than one at a time. Our infrastructure runs through data centres located outside CLOUD Act jurisdiction, which is intended to support data sovereignty for defence and high-threat environments. And we expect our blockchain infrastructure experience to support tamper-proof, immutable data pathways for sensor readings, designed to give a verifiable chain of custody for detection data.1
That last one sounds technical and dull. It is neither. If a detection reading is ever going to support a decision that ends up in front of a court, a regulator or a commander, somebody has to be able to prove the reading was not altered between the device and the screen.
The capability gap is real, it is documented, and it sits inside a very large and rapidly growing budget line. We have signed an agreement intended to address it.
Where this goes
Sight is largely solved. Sound is largely solved. The remaining sense is the one that detects the things you never see coming, and it is the one that has resisted every attempt to brute-force it with better materials alone.
If it does get solved, and I think it will, my strong suspicion is that it will not be solved by whoever builds the most sensitive material. It will be solved by whoever builds the layer that reads what the material picks up, quickly enough and reliably enough to be trusted with a decision that matters.
The sensor is not the product. The interpretation is.
It is why we signed the agreement, and it is why I am not going to explain the rest of it here.
Jason Dussault is Chief Executive Officer and Co-Founder of Intellistake Technologies Corp. (CSE: ISTK | OTCQB: ISTKF | FSE: E41). More information is available at intellistake.com Sources
1. Intellistake Technologies Corp., news release, August 4, 2026, "Intellistake Signs C$17 Million Defense AI Acquisition Agreement for NanoAi Technologies."
https://www.newswire.ca/news-releases/intellistake-signs-c-17-million-defense-ai-acquisition-agreement-for-nanoai-technologies-880607578.html2. Minami K., "Nanomechanical Sensors for Gas Detection towards Artificial Olfaction," Biosensors, 2022, 12(4), 256:
https://doi.org/10.3390/bios120402563. Horvat T., Pehnec G., Jakovljevic I., "Volatile Organic Compounds in the Atmosphere," Toxics, 2025, 13(5), 344: https://doi.org/10.3390/toxics13050344 ; and US Environmental Protection Agency, "Volatile Organic Compounds' Impact on Indoor Air Quality":
https://www.epa.gov/indoor-air-quality-iaq/volatile-organic-compounds-impact-indoor-air-quality 4. US Environmental Protection Agency, "Technical Overview of Volatile Organic Compounds":
https://www.epa.gov/indoor-air-quality-iaq/technical-overview-volatile-organic-compounds 5. The Nobel Prize in Physiology or Medicine 2004, Richard Axel and Linda B. Buck, "for their discoveries of odorant receptors and the organization of the olfactory system":
https://www.nobelprize.org/prizes/medicine/2004/summary/ 6. Olender T., Lancet D., Nebert D.W., "Update on the olfactory receptor (OR) gene superfamily," Human Genomics, 2008, 3(1), 87-97:
https://doi.org/10.1186/1479-7364-3-1-87 7. Covington J.A., Marco S., Persaud K.C., Schiffman S.S., Nagle H.T., "Artificial Olfaction in the 21st Century," IEEE Sensors Journal, 2021, 21(11): https://doi.org/10.1109/JSEN.2021.3076412
8. Angle C., Waggoner L.P., Ferrando A., Haney P., Passler T., "Canine Detection of the Volatilome: A Review of Implications for Pathogen and Disease Detection," Frontiers in Veterinary Science, 2016, 3:47:
https://doi.org/10.3389/fvets.2016.00047 9. Farr B.D., Otto C.M., Szymczak J.E., "Expert Perspectives on the Performance of Explosive Detection Canines: Performance Degrading Factors," Animals, 2021, 11(7):
https://pmc.ncbi.nlm.nih.gov/articles/PMC8300196/ 10. Lee B.K. et al., "A principal odor map unifies diverse tasks in olfactory perception," Science, 2023, 381, 999-1006:
https://doi.org/10.1126/science.ade4401 11. University of Pittsburgh, Pittwire, "Chemical fentanyl sensor a star of research group," May 1, 2024:
https://www.pittwire.pitt.edu/pittwire/features-articles/chemical-fentanyl-sensor-star-research-group ; underlying paper, Shao W., Sorescu D.C., Liu Z., Star A., Small, 2024, 20(35):
https://doi.org/10.1002/smll.20231183512. Stockholm International Peace Research Institute, "Global military spending rise continues as European and Asian expenditures surge," April 27, 2026:
https://www.sipri.org/media/press-release/2026/global-military-spending-rise-continues-european-and-asian-expenditures-surge 13. NATO, The Hague Summit Declaration, June 25, 2025:
https://www.nato.int/cps/en/natohq/official_texts_236705.htm 14. United States Department of War, Office of the Under Secretary of Defense (Comptroller), "FY2026 Budget Request":
https://comptroller.war.gov/Portals/45/Documents/defbudget/FY2026/FY2026_Budget_Request.pdf 15. The White House, "Rebuilding Our Military" fact sheet:
https://www.whitehouse.gov/wp-content/uploads/2026/04/rebuilding-our-military-fact-sheet.pdf 16.The Business Research Company via Research and Markets, "Artificial Intelligence and Analytics in Defence," January 2026:
https://www.researchandmarkets.com/reports/6215080/artificial-intelligence-analytics-in-defence 17. DataM Intelligence, "Military AI Market":
https://www.datamintelligence.com/research-report/military-ai-market 18. Unmanned Airspace, "Global spending on counter-UAS systems to reach USD12.6 billion this year," June 3, 2026:
https://www.unmannedairspace.info/counter-uas-systems-and-policies/global-spending-on-counter-uas-systems-reach-usd12-6-billion-this-year/ 19. Al Jazeera, "Denmark shuts second airport in a week, more unidentified drones spotted," September 25, 2025:
https://www.aljazeera.com/news/2025/9/25/denmark-shuts-second-airport-in-a-week-more-unidentified-drones-spotted 20. VRT NWS, "Air traffic halted last night due to drones at Brussels Airport," November 5, 2025:
https://www.vrt.be/vrtnws/en/2025/11/05/air-traffic-halted-last-night-due-to-drones-at-brussels-airport/ 21. Stars and Stripes, July 2, 2026, reporting research published by the International Institute for Strategic Studies:
https://www.stripes.com/theaters/europe/2026-07-02/russian-drones-nuclear-bases-nato-22148620.html 22. US Centers for Disease Control and Prevention, National Center for Health Statistics, "U.S. Overdose Deaths Decrease in 2025," May 13, 2026:
https://www.cdc.gov/nchs/pressroom/releases/20260513.html 23. US Department of Homeland Security, "CBP Seizes More Than 100 Million Fentanyl Doses Along Southwest Border in 2026," May 15, 2026:
https://www.dhs.gov/news/2026/05/15/cbp-seizes-more-100-million-fentanyl-doses-along-southwest-border-2026 Disclaimer
Completion of the NanoAi Transaction remains subject to customary conditions including completion of satisfactory due diligence (including verifying title to the intellectual property of NanoAi), completion of the audit of financial statements of NanoAi, the Company maintaining a minimum cash balance of $2 million and no objection from the Canadian Securities Exchange. Closing is targeted within 60 days thereafter, subject to satisfaction of closing conditions in the Definitive Agreement.
This report contains "forward-looking information" concerning anticipated developments and events related to the Company that may occur in the future. Forward looking information contained in this report includes, but is not limited to, all statements in respect of the Company's growth and development, expectations regarding prediction market growth, the operations and business segments of the Company and NanoAi, the functionality of the Company’s software, and its benefits, the details of the proposed acquisition of NanoAi, the conditions to completion of the proposed acquisition of NanoAi, the benefits of the acquisition of NanoAi, the business model of NanoAi and future potential recurring revenues, the synergies between NanoAi and the Company, and bridging the gap between emerging decentralized networks and real-world industry adoption.
In certain cases, forward-looking information can be identified by the use of words such as "expects", "intends", "anticipates" or variations of such words and phrases or state that certain actions, events or results "may", "would", or "might" suggesting future outcomes, or other expectations, assumptions, intentions or statements about future events or performance. Forward-looking information contained in this report is based on certain assumptions regarding, among other things, the Company will continue to have access to financing until it achieves profitability; the Company and NanoAi satisfy all conditions necessary to close the proposed transaction; the technology and blockchain industries in which the Company intends to focus its business in will grow at the rate and in the manner expected; the ability to attract qualified personnel; the success of market initiatives and the ability to grow brand awareness; the ability to distribute Company's services; the Company creates strategies to mitigate risks associated with cryptocurrency price fluctuations; the Company remains compliant with all applicable laws and securities regulations and applicable licensing requirements; the Company engages and collaborates with local experts, as necessary, to address jurisdiction-specific matters and ensures compliance with foreign regulations to avoid penalties; the Company addresses any potential cybersecurity threats promptly and effectively; the ability of the Company to develop its technology, acquire customers and have revenue; the ability to successfully deploy the new business strategy as a result of the change of business. While the Company considers these assumptions to be reasonable, they may be incorrect.
Forward looking information involves known and unknown risks, uncertainties and other factors which may cause the actual results to be materially different from any future results expressed by the forward-looking information. Such factors include risks related to general business, economic and social uncertainties; failure of the Company and NanoAi to satisfy all conditions necessary to close the proposed transaction; failure to raise the capital necessary to fund its operations; inability to create strategies to mitigate the risks associated with cryptocurrency price fluctuations; the costs of regulation in the digital asset industries increase to the extent that the Company is no longer generating sufficient returns for shareholders; failure to promptly and effectively address cybersecurity threats; insufficient resources to maintain its operations on a competitive basis; and the actual costs, timing and future plans differs expectations; legislative, environmental and other judicial, regulatory, political and competitive developments; the inherent risks involved in the cryptocurrency and general securities markets; the Company may not be able to profitably liquidate its current digital currency inventory, or at all; a decline in digital currency prices may have a significant negative impact on the Company's operations; the Company's success may depend on the continued involvement of key personnel, including advisors, whose involvement cannot be guaranteed; institutional adoption of decentralized AI infrastructure remains uncertain and may not occur at the pace or scale anticipated; evolving regulatory frameworks, including those related to AI (such as Canada's proposed Artificial Intelligence and Data Act) and prediction markets, may impose additional compliance burdens or restrict certain business activities; valuation figures are based on publicly available market data and internal assessments at the time of the referenced transactions and may not reflect current or future valuations; the volatility of digital currency prices; the inherent uncertainty of cost estimates and the potential for unexpected costs and expenses, currency fluctuations; regulatory restrictions, liability, competition, loss of key employees and other related risks and uncertainties; delay or failure to receive regulatory approvals; failure to attract qualified personnel, labour disputes; and the additional risks identified in the "Risk Factors" section of the Company's filings with applicable Canadian securities regulators.
Although the Company has attempted to identify factors that could cause actual results to differ materially from those described in forward-looking information, there may be other factors that cause results not to be as anticipated. Readers should not place undue reliance on forward-looking information. The forward-looking information is made as of the date of this report. Except as required by applicable securities laws, the Company does not undertake any obligation to publicly update forward-looking information.
