In a world reeling from massive technology-driven systems change, investment bubbles in new new things that are billed to be Holy Grail solutions to wicked problems are both features and bugs.
This is certainly true of the outsized investments in the race to build Artificial General Intelligence (AGI), a type of artificial intelligence (AI) that matches or surpasses human capabilities across a wide range of cognitive functions.
It is also true of the green energy equivalent, Green hydrogen.
Green hydrogen is produced from water through a process powered by renewable energy generation. It has often been called the Swiss Army knife of renewables with pundits predicting it can be turned to almost any purpose.
When technology sounds too good to be true, it usually is. But not all tech-led bubbles are the same, and the trick is to make your bets on the one which is more feature, than bug.
Australia’s Future Made in Australia agenda bet big on the latter while ignoring the former. And it is also revealing about who in Australia influences key decisions and investment priorities.
First, let’s look at the AI bubble.
There is a lot we should take away from the recent market meltdown, which saw the share price of the Magnificent Seven group of tech companies that are also the biggest investors in artificial intelligence lose more than US$1.5 trillion in market value.
Last week’s weaker than expected jobs report in the United States triggered fears the superpower was heading into recession. That coincided with the unwinding of the Japanese carry trade, as well as increased geopolitical instability and heightened conflict between Israel and Iran.
All converged in a perfect storm of weaker earnings reports from several of the big tech giants to create market volatility globally.
The US recession fears have cooled in recent days. But the skittishness about the viability of AI still dominates tech and financial news.
But while the quest to achieve Artificial General Intelligence (AGI) may be a long time coming, if ever, the torrent of reports predicting the bursting of the AI bubble is premature, if not entirely misplaced.
We are only two years into the AI investment boom which began with characteristic irrational exuberance with the launch of OpenAI’s ChatGPT in November 2022 and powered the share prices of the Big Tech giants to new heights.
And now, the tech giants’ huge capital expenditures in AI chips and infrastructure like data centres and increasing fears of weak customer demand have analysts concluding AI has hit a roadblock.
The primary takeaway is that AI is costing too much, while enterprise customers are having issues implementing the enabling technology or are not seeing the productivity gains when they do.
The much-hyped AI era has always had its sceptics and on the surface, the nervousness about AI seems warranted.
Microsoft, Amazon, Google and Facebook parent company, Meta Platforms have seen their stocks slide an average of 11 per cent from their record highs in early July.
The four tech giants are dumping record amounts of capex into AI, which hasn’t yet grown into a significant enough revenue source to merit disclosure by any of them.
The only clear winner to date is AI chip maker, Nvidia which seen a $US900 billion drop in its share market value in the last two months.
And early last month, a widely read Goldman Sachs report: Gen AI: Too Much Spend, Tool Little Benefit? poured an avalanche of ice onto the AI hype cycle.
This was despite it being only a month after another research report from the same investment giant told the world AI would boost GDP and raise productivity over the next decade.
In the latest report, Goldman’s highly respected head of Global Equity Research, Jim Covello argued that to earn an adequate return on the more than US$1 trillion estimated cost of developing and running AI technology, it must be able to solve complex problems which, he says, it isn’t built to do.
He points out that truly life-changing inventions like the internet enabled low-cost solutions to disrupt high-cost solutions, even in its infancy.
AI is not doing anything like that yet, the report concluded, with a caveat that it’s still early days for the transformative platform technology.
But it is still very early days indeed for AI and the spending wars are far from over. And although immature and hard to price and implement, AI-enabled software and applications are popping up everywhere, fueled by an ambitious and growing cohort of startups looking to ride the wave.
Even with meagre revenues from AI-enabled products, the tech industry’s biggest companies have made it clear over the last week that they have no intention of throttling back their spending on artificial intelligence.
They know that AI is much more an enterprise play than a consumer one. And any transformative enabling or platform technology takes longer to filter through to corporate enterprise customers.
In just the last quarter, Apple, Amazon, Meta, Microsoft and Google’s parent company Alphabet spent a combined US$59 billion on capital expenses, 63 per cent more than a year earlier and 161 per cent more than four years ago.
Only Apple has not dramatically increased spending because it does not build the most advanced AI systems itself.
Back in 1995, at the birth of the commercial internet era which built a mother of a bubble that burst in 2001 before surviving companies (Google, Facebook, Amazon) regrouped and won the world, Microsoft founder Bill Gates told the author of this column that businesses and policymakers almost always overestimate the impact of new technology over two years, while dramatically underestimating its impact over ten years.
It was true for the internet and the era of digital commerce it ushered in, and it is likely also true for artificial intelligence application development and adoption. The tech giants have mountains of cash to invest in the long game and know the price of not winning it.
Meanwhile, the Productivity Commission’s assessment of the AI development opportunity in Australia was lower down the priority list in the face of the huge capital outlays and accelerated activity by the US tech giants. Best to craft responsible AI regulation, it suggested, and use the AI created by other countries.
Ignoring the new enabling tech is shortsighted and ultimately dangerous to any country’s prospects for economic growth and productivity in the next phase of the 21st century.
While the Australian Labor government and large businesses seem to want to ignore the AI revolution, it was a different story for green hydrogen – a volatile and unproven technology with many sceptics the world over.
Hydrogen and its derivatives, particularly ammonia and methanol, were seen as the great green hope for the green energy transition and the FMiA agenda.
Encouraged by highly credentialed lead advocates such as former chief scientist of Australia, Dr Alan Finkel and iron ore billionaire Twiggy Forrest, both the previous Coalition government and the current Labor government fully embraced the green hydrogen dream.
Dr Finkel was commissioned by the Coalition government to produce a road map for creating the nation’s green hydrogen industry which was released in 2019. When Labor came to power, it commissioned a review of the green hydrogen industry roadmap but remained a firm believer in it.
In the recent federal budget, Labor introduced some $4 billion in tax credits to accelerate a domestic green hydrogen industry with the flagship Hydrogen Headstart program.
“Green hydrogen is going to have its day. This is multi-generational change. This is forever change,” crowed Dr Forrest, who called hydrogen sceptics – including Elon Musk – “Muppets”.
Dr Forrest invested over $1 billion dollars in projects, set a target of producing 15 million tonnes of hydrogen by 2030, and announced partnerships with various state and federal governments with the R&D transition
But two weeks ago, even as Australia’s most ardent backer of green hydrogen which he wanted to export to the world, Twiggy Forrest announced a massive scaling back of his green hydrogen projects.
And although no one is saying it out loud, the general consensus is that the green hydrogen bubble has burst.
Dr Finkel, who is the only person in Australia who drives a hydrogen-fueled car, and who once said Australia could become a global leader in “shipping sunshine”, says his support for the clean energy source has “evolved”.
He acknowledges that a lot of the uses he foresaw for hydrogen have proven impractical, primarily due to high energy prices to produce it.
For instance, he says it’s difficult to see green hydrogen replacing gas for heating homes (which was number one of six great benefits of the much-hyped clean energy source in the 2019 strategy.)
Instead, Dr Finkel says: “The main role for green hydrogen production in Australia will be as a chemical to produce decarbonised products for export, such as green iron, green ammonia, green fertiliser, green jet fuels and renewable diesel (especially to Japan and South Korea)
“Using renewable electricity to make green hydrogen, then store it, then feed it to a converted gas generator is expensive and inefficient.”
Where this leaves the government’s hydrogen hopes for Australia’s FMiA mission is unclear.
Federal Energy and Climate Minister Chris Bowen recently reaffirmed the commitment to hydrogen saying, “reports of the death of the green hydrogen industry are greatly exaggerated.”
“We’ve had one project by one company deprioritised. That does not mean that hundreds of other projects under development have been deprioritised.”
He added that more than $200 billion worth of investment projects are in the pipeline.
Still, the Opposition, which is committing to building nuclear power plants by 2037, pounced on the Forrest project cancellation and there is little doubt that there are many more headwinds for hydrogen, nuclear and renewables projects to come. All will put the energy debate in the lead up to the next federal election, and the high energy costs businesses and consumers are already reeling from, take centre stage.
Meanwhile, one ASX-listed hydrogen technology startup seems poised for growth. Perth-based Hazer Group has taken far more pragmatic approach to the risky and unproven technology by being squarely focused on “clean” hydrogen as opposed to the costly and unproven “green” hydrogen.
It’s chief executive, Glenn Corrie says his technology can deliver gas at one-fifth the costs of renewables-based hydrogen.
“Green is far too expensive, and too far out into the future,” Mr Corrie said. He says the Hazer technology could provide hydrogen buyers with gas at between US$2 and US$3 a kilogram, compared to north of US$8 per kilogram for green hydrogen.
The Hazer process, which is IP protected, enables the conversion of methane-rich gas and similar feedstocks into hydrogen and carbon in the form of high-quality graphite using iron ore as a catalyst.
Hazer stands for Hydrogen and Zero Emissions Research, and Mr Corrie says the 15-year-old company is now moving into the commercialisation phase of the business.
Its commercial demonstration plant (CDR) has achieved more than 100 hours of continuous operation – a significant milestone representing a Technology Readiness Level of 7 out of 9 (a NASA standard). It represents the fifth consecutive scaling up of the process.
Hazer has spent $105 million developing the technology, utilising the R&D Tax Incentive rebate and recently raised another $15 million for an extended runway to meet further milestones.
Corrie emphasizes that what differentiates Hazer from other listed Hydrogen hopefuls in Australia is that it’s a technology company, not a project developer, so now has an accelerated path to commercialisation.
He has signed up three Tier One global customers for industrial-scale projects, including a licensing agreement with the Vancouver-based LNG energy company, Fortis BC, which is twice the size of Australia’s Origin Energy. The other foundation projects are Chuba Electric Power (Japan) and the French company Engie.
“No red flags,” says Mr Corrie. “The tech is proven. What we are now doing is demonstrating it and preparing it for commercialisation.
He adds that there are multiple discussions with oil and gas companies, steel and cement makers and petrochemical plants that could make for a busy and cash-rich twelve months ahead.
Do you know more? Contact James Riley via Email.