Small Caps | May 02 2023
This story features BRAINCHIP HOLDINGS LIMITED, and other companies. For more info SHARE ANALYSIS: BRN
By Tim Boreham
The artificial intelligence (AI) revolution is shaping up as the most profound technology development since the invention of the internet – and it’s naturally one that investors will be discussing with their financial advisors (or, ironically, Chatbot).
Currently, the ASX-listed AI sector can be split between tech providers developing AI tools and non-tech companies – anything from miners to bankers – using AI to improve their existing business.
Plenty of hype and trumped-up claims are emerging, so investors need their old-fashioned BS detectors on a high setting. Even ChatGPT cautions that AI is a relatively new field and “subject to rapid changes and technological disruption, which could impact the performance of AI companies in the future.”
One intriguing AI pioneer is Brainchip Holdings ((BRN)) which is commercialising Akida, a neuromorphic process to provide and improve data to develop machine learning and AI products.
It’s also got something to do with ‘spiking neural networks’ and good on them – with a $680m market cap investors are convinced about the prospects of the company which is generating some revenue, but also deep losses.
Valued at $1.2bn, Weebit Nano ((WBT)) develops “resistive Random-Access Memory technologies which are a specialised form of non-volatile memory for the semiconductor industry.”
Weebit has just raised $60m via a placement and “upsized and scaled-back” share purchase plan, so once again there’s no shortage of investor confidence.
With a $340m market cap, Nuix (NXL) provides investigative analytics and intelligence software with a charter of “finding truth in a digital world”.
The truth is out there – but in the zeros and ones rather than at a Roswell UFO convention.
At the sub $100m market cap end, how do investors leverage a theme that – as with the dial-up internet – is likely to evolve over time?
The company behind much of the broadcast captioning both here and the US, AI Media ((AIM)) uses extensive AI to enable seamless translation, even if the speaker is a Russian crime boss with a dodgy accent.
The company is transitioning customers from its old iCap product to its AI-driven platform called Lexi. The key difference? Lexi doesn’t need a human with a familiar voice to repeat the words into a speech recognition program.
According to founder and CEO Abrahams, the way audio is delivered into a speech recognition program is a “massive determinant” of the accuracy of the caption. “We will mic up a commentators booth with isolated audio and no background noise,” he says.
The company has dozens of automated speech recognition engines, selected according to the accent or language and the type of setting (such as a horse race versus a news broadcast).
AI Media reported flat revenue of $29.7m for the half year to December 2022. But earnings before interest, tax depreciation and amortisation (ebitda) was $1.4m, partly the result of upgrading customers to Lexi.
AI Media’s customers include Foxtel, Al Jazeera, SBS, Google, Channel Seven and Major League Soccer.
The company also won contracts for the cities of San Francisco, Austin and Baltimore and is eyeing expanded opportunities in parliaments and court rooms.
On a same-but-different note, Straker Translations ((STG)) is using the most advanced AI to clean up the hitherto error-ridden and unintentionally amusing machine-driven efforts that are still evident on Google Translate.
Rather than slavishly and literally translating word by word, AI tools can understand tone and complex sentence structures.
They can even understand slang and jokes, although they are not at the point of being able to crack one. And we’re sure the Kiwi-based Straker’s platform can differentiate its Toms from its Tums.
By any translation, Straker has been having a tough time because of sluggish global economic conditions. Last year the company launched cost cutting drive, eliminating -15% of its global workforce (now 225 full-time equivalents).
The results of the tidy-up are becoming apparent, with the company last week reporting fourth (March) quarter positive cash flow of NZ$1.7m. Full year revenue was 6% higher at NZ$59.4m, with NZ$1.4m of ‘adjusted ebitda’ compared with NZ$200,000 previously.
Another exemplar of the practical uses of AI is the $10m market cap security monitoring house Icetana (ICE), which uses machine learning to filter out routine motion on a camera image. As a result, only the suspicious activities are brought to the attention of the monitors.
The company’s guiding concept principle is YOLO – as in ‘you only look once’.
Icetana has contracts covering 60 sites on five continents. Last month the company renewed a contract with a Middle East shopping centre owner, covering 16 malls and 8000 cameras for the next three years.
Icetana last week reported a March (third) quarter receipts of $406,000 and cash outflows of -$605,000, with a meagre cash balance of $1.4m. Cut another way, annual recurring revenue of $1.6m was 12% stronger than a year previously.
Meanwhile, the $80m cap Acusensus ((ACE)) operates road traffic cameras to detect illegal mobile usage and non-seat belt wearing.
While the company’s smarts relate more to the type and positioning of the cameras, it is harnessing AI to widen its applications to other road safety measures. One example is the early, real-time detection of drivers likely to be drug affected, based on tell-tale signs such as lane swerving.
Acusensus has reported March (third) quarter revenue of $10.9m, 12.5% higher year on year with positive cash flow of $2.6m.
Management has upped its full-year revenue forecast to “at least” $40m, which compares with the estimated $37m ahead of last January’s IPO. The number is also 39% up on the previous year’s turnover.
The Acusensus units add to the estimated one billion security cameras in the world – or more than one for every eight inhabitants of the planet.
That sounds pretty creepy, but victims of muggings or road accidents might think differently.
Disclaimer (courtesy of Chatbot): AI is a relatively new field and is subject to rapid changes and technological disruption, which could impact the performance of AI companies in the future. Additionally, the stock market can be unpredictable, and it's always possible that a company's performance may not meet expectations.
Find out why FNArena subscribers like the service so much: "Your Feedback (Thank You)" – Warning this story contains unashamedly positive feedback on the service provided.
FNArena is proud about its track record and past achievements: Ten Years On