Griffin on Tech: An AI nothingburger and X chief groks out
I guess, as the AI Forum’s Madeline Newman put it this week, the Government’s new AI strategy is at the very least a “conversation starter”.
However, the AI conversation has been ongoing in New Zealand for some time. I’m just not sure that senior government ministers have been listening.
New Zealand was late to the party in developing a plan for exploiting the opportunities and managing the risks associated with the emergence of AI, which has gained significant momentum in the most recent era of generative AI. The benefit of being last is that you can look at what everyone else is doing and cherry-pick the best bits.
But it's not clear that we are even doing that. We seem to have decided it's all too hard and expensive. The tone of feedback I’ve read from industry experts suggests the AI strategy is a major missed opportunity, given that AI supposedly has the potential to add $76 billion to the economy by 2038, according to the strategy document.
Little that’s new
Massey University’s Dr Collin Bjork perhaps summed it up best: “On one level, it’s a nothing burger because the government promises very little in the way of new policies or investments around AI. Instead, they largely point to existing investments and policies that were already in the works and that they believe will help with AI uptake”.
I agree entirely. There wasn’t a single thing in the strategy that suggested some evolving thinking about approaching AI in the context of Aotearoa. The previously announced advanced technology public research organisation, which is still to take shape but has the mandate of focussing on AI and other advanced technologies, at least points to a dedicated effort in this space. But it will take a long time, and successful execution, for that effort to pay off.
Speaking of space, look at the level of policy attention and resourcing the Government has devoted to the space sector, including most recently, the Royal New Zealand Air Force’s establishment of a dedicated 12-person space unit. Our space sector is doing very well, but the overall economic impact of AI will likely dwarf what our space companies generate in export earnings.
Despite that, there’s essentially nothing in the strategy dedicated to upskilling the workforce, planning for the displacement of jobs as AI agents and other more advanced forms of AI do more white collar work. The Government knows that small and medium-sized businesses, making up the vast bulk of our economy, are dragging the chain on AI adoption.
Aussie on the front foot
But unlike Australia, which set up the National AI Centre with the explicit purpose of driving AI adoption. Unlocking the potential of AI on an accelerated timeline needs some sort of dedicated push, coordinated and funded by the government with clear targets and milestones. Instead we have a version of the digital strategy developed on Labour’s watch, which was generic and vague and gathered dust.
Then there are the ethical questions. As Victoria University’s Dr Andrew Lensen put it: “The Strategy suggests that new legislation is unnecessary, which I, and many other AI researchers, disagree with,” he told the Science Media Centre.
“Having ‘Principles’ is not nearly sufficient to reduce AI-induced harm, bias, and inequity—we need clear legislation and well-resourced enforcement mechanisms to ensure AI does not further harm New Zealanders.”
Where’s the upside?
A nothing burger indeed. So where’s the upside? This Government is laser-focussed on economic development. As with the space sector, where it sees an opportunity for quick economic wins, it gets interested quickly, does its homework, and then gets serious on the policy front.
It has talked up Halter as an example of a New Zealand tech success that is leveraging AI to great effect. That’s the language the coalition government talks. Businesses have been given the signal to get on with adopting AI (but not developing it, because that’s too complicated and expensive).
The light-touch regulatory regime will facilitate that, even if it may actually hinder tech exporters who are developing for markets like the European Union, and Australia, which have or will soon have dedicated AI regulation. It’s now up to businesses and the tech sector to show the real productivity gains and economic wins that can come from putting AI to work. Once this collection of politicians sees the momentum building, they are more likely to get in behind it with enabling policies and tangible initiatives.
When the bots write the headlines
So Linda Yaccarino has left the X building, presumably with her dignity, if not her blue checkmark, still intact. The timing? Impeccable. Just as X’s Grok AI chatbot was busy spewing antisemitic nonsense and nostalgia for Hitler, Yaccarino decided she’d had enough of playing chief firefighter for Elon Musk’s latest dumpster inferno.
Who could blame her? When your company’s “cutting-edge” AI starts channelling 4chan rather than Alan Turing, maybe it’s time to update your résumé.
Grok’s hateful outbursts are a flashing red warning that the race to build “edgy” AI without ethical brakes is a disaster in the making. Responsibility can’t be open-sourced, delegated to the next update, or redefined based on the whims of Elon Musk.
The tech world needs less bravado and more backbone, starting with ethical guardrails that actually work backed by legislation that actually works.
Agents of change
Finally, I really enjoyed my podcast discussion this week with J.D. Trask, the Wellington co-founder of Raygun who has launched a new venture, Autohive, a platform for SMEs seeking to build and run AI agents. A few salient points J.D. makes which are useful in approaching AI:
What's one monotonous task that takes you one hour or more a week, or month? Use AI to try and automate it.
Remember the Jeff Bezos mantra, which came from him realising that the internet would change so much about the way we live and work. Focus on the things that won't change. For Bezos, it was the fact that people would still want to read books, they'd just want them delivered digitally via the internet.
Revisit the Bill Gates Internet Tidal Wave memo, which he published on May 26, 1995, 30 years ago. We are at another of those inflection points. AI is a similarly pivotal waypoint in the evolution of technology.
Everyone is talking about AI agents, especially the Big Tech vendors. But AI agents, says J.D., are like an "infinite box of Lego". The pieces are scattered all over the table, but it takes the right person to put them together.
Listen to the Business of Tech podcast interview with J.D. Trask here.
