‘Tokenmaxxing’ is dead: How the hardware shortage is blowing up AI budgets
Google capped access to its Gemini AI models, signaling the decline of 'tokenmaxxing' as AI costs soar amid a global hardware shortage. Meta, a major AI customer, failed to secure desired capacity, highlighting the severity of constraints. An Australian study shows one in three companies exceeded AI budgets, prompting a reevaluation of spending strategies.
Take it further — get the full app and never miss a moment of what's happening in Australia.
This publisher's site can't be shown here due to their security settings.
Open full article →No source link available for this article.
Published 29 Jun 2026, 06:58 UTC · Updated 29 Jun 2026, 07:00 UTC
Summary by OZbrief Editorial. Original report: SMH. Editorial policy · Corrections
Trending
- 'It's who we are': Albanese issues impassioned defence of multiculturalism
- Cody Gakpo and partner Noa van der Bi announce miscarriage of son
- Court returns proceeds of house sale seized from convicted paedophile
- Abigail Velez apologizes for controversial comment about Bosnia during broadcast
- Former CFMEU chief pleads for drug trafficker tied to Big Build corruption
- Katrina Kroetch showcases women's magic talent at Melbourne Magic Festival



