Dario Amodei lands in Canberra, red carpet out, politicians swooning.
Anthropic’s boss—putative good cop in the AI oligarchy—spins his yarn. Good AI. Safe AI. The kind that boosts economies without, you know, turning us all into gig serfs. Australia’s government laps it up, signs deals, releases datacentre principles timed to perfection. Coincidence? Please.
Here’s the thing. Amodei bolted OpenAI over safety lapses. Pens thoughtful essays on tech perils. Ditched Trump-era spy dreams. Sounds noble. Almost.
But Toby Walsh, AI prof at UNSW, nails it:
there is no “good AI” because AI is both good and bad.
Spot on. These models guzzle energy, train on pilfered creator work—Anthropic settled a $1.5bn authors’ lawsuit, remember?—and Amodei himself predicts half of entry-level white-collar jobs vaporized. Yet on he charges.
Is Anthropic’s ‘Progressive AI’ Just Corporate Spin?
Look, I’ve marched for progress my whole life. Union dad. Dismissal kid. Hawke era. Decades in the trenches—unions, parties, climate, First Nations. Wore the tees. Believed the arc bends toward justice.
Social dems would steer the ship. Redistribute. Safety nets. Tax the rich. Government as moral compass.
Then bam. Cultural bubbles. Economic seduction by right-wing vibes. Tech’s “progress” steamroller.
AI’s the latest wolf in progressive clothing. Government all-in on productivity fairy tales—industry models, naturally. Unions grumble. Artists rage. Parents fret screen-zombie kids. But hey, Anthropic MOU promises safety tracking, vibrant ecosystems. Ticks boxes. Feels good.
Doesn’t it?
Amodei regales PM, suits, glitterati. Transforms economy, he says. Signs first on datacentres. Shill perfection.
From cheap seats: this ain’t progress. It’s accelerationism with a smiley face.
And my unique twist? This mirrors the Luddites smashing looms in 1811—not anti-progress, but pro-humanity. They knew machines don’t vote, pay taxes, or care about fairness. AI overlords like Amodei? Same blind spot. Bold prediction: without iron-fisted global regs, AI widens the chasm—top 1% coders and CEOs feast, rest scrape Uber scraps. Progressivism dies not with a bang, but a bankrupt UBI pilot.
Why Does ‘Safe AI’ Still Destroy Jobs?
Amodei admits it. Up to 50% white-collar entry gigs gone. Poof.
Yet Canberra cheers. Productivity! Growth! As if GDP trumps dignity.
Energy hog. Stolen data. Machines replacing thinkers. Good and bad baked in, per Walsh.
Government’s contradiction city. Core voters—unions, creatives—sidelined for tech bribes. One Nation laughs, turbocharges backlash. Identity politics? Sure. But economic gut-punch from AI? That’s the real populist fuel.
Progressive governments? Seduced. Forgot private sector’s not our pal. Essential services? Nah, let algorithms deliver.
I’ve defined “progressive” loosely. Momentum toward justice. Hegel. MLK.
No more. Tech shoves its vision—unilateral, unvoted. Waves of so-called progress birthing Trumps, One Nations.
Hit brakes.
But wait—Amodei’s not worst. OpenAI ditched safety-first. He split. Essays on fears. No spy tech for tyrants.
Commendable. Ish.
Still builds the beast. Settles creator suits. Predicts mass unemployment. Carries on.
No good AI. Just AI. Our job: tame it before it tames us.
Will AI Kill the Progressive Dream?
Australia’s all-in. Datacentres. Frontier tracking. Ecosystems.
Manna for contradiction-plagued pollies.
But artists? Outraged—rightly. Their work, scraped.
Parents? Kid futures automated away.
Unions? Silent-ish, for now.
Me? Lifelong progressive, calling bullshit. World won’t auto-bend to justice. Not with unchecked AI.
Redistribute what? Tax whom? When coders own the machines?
Unique parallel: 1930s auto boom. Unions fought, won shares of pie. AI? No one’s fighting. Yet.
Prediction: populist surge if we don’t. Liberals topple. Progressives? Blame themselves.
Brakes. Now.
Short version: Amodei’s charm offensive masks the grind. Good intentions pave data hells.
Wake up.
🧬 Related Insights
- Read more: OpenAI Buys TBPN: Narrative Control Masquerading as Dialogue
- Read more: 2026’s Open LLM Avalanche: 10 Architectures That Promise More Than They Deliver
Frequently Asked Questions
What did Anthropic’s CEO promise Australia?
Dario Amodei pitched economy-boosting, safe AI, signed datacentre principles, and inked an MOU for safety tracking and local ecosystems.
Is Anthropic’s AI really ‘good’ and safe?
No—it’s dual-use: innovative but energy-intensive, trained on unlicensed data, and set to wipe entry-level jobs, per critics like Toby Walsh.
Will AI destroy half of white-collar jobs?
Amodei predicts up to 50% of entry-level roles gone; without regulation, it’s a progressive nightmare for workers.
