Sep 15, 2026 · 11 min

AI safety pressure collides with relentless investment

OpenAI Says It’s Working With Anthropic, Google on AI Safety

The episode examines whether voluntary safeguards can keep pace with AI spending, political scrutiny, and divergent approaches to adoption in the United States and Europe.

3 key takeaways

1AI companies may pursue voluntary frontier-model guidelines before Congress can agree on binding legislation.
2Investment is likely to continue, even if development slows, by shifting toward existing models and new applications.
3Europe’s AI adoption remains healthy but more cautious, with data-center expansion and environmental effects under closer scrutiny.

Don't miss

Matthew Bloxham explains why AI spending could continue even if companies agree to slow frontier-model development.

The brief

Political pressure is rising as the United States weighs AI’s economic and employment effects against strategic competition with China and calls for continued investment.

Matthew Bloxham says companies are likely to pursue voluntary frontier-AI guidelines first, while congressional agreement on eventual U.S. legislation remains difficult.

Even a slowdown in frontier-model development would not necessarily curb spending: investment could shift toward improving existing models, differentiation, and new applications.

OpenAI and Anthropic’s potential IPOs depend not only on company progress, including Anthropic’s profitability, but also on broader stock-market conditions.

Europe offers a contrasting signal: adoption is relatively healthy but more cautious, with greater attention to data-center deployment and environmental effects.

What was said on this episode

17 statements · 11 positive · 2 negative · 2 mixed · 2 neutral

Paul Sweeneyon US-China AI competitionNegative2:23
Reducing AI investment risks the US and West falling behind China
“this is kind of competitive race, China versus the West, China versus the US. So to the extent you do maybe pull back for whatever reason, you do run the risk of maybe falling behind.”
Listen at 2:23
Matthew Bloxhamon voluntary AI industry guidelinesPositive3:16
Voluntary AI industry guidelines are the best short-term regulatory outcome
“in the short term, The best we can hope for in terms of clarity would be some kind of voluntary industry guidelines.”
Listen at 3:16
Matthew Bloxhamon AI industry guidelinesPositive3:18
Voluntary AI industry guidelines are the best near-term regulatory outcome
“The best we can hope for in terms of clarity would be some kind of voluntary industry guidelines.”
Listen at 3:18
Matthew Bloxhamon U.S. AI regulationNeutral3:43
US legislation may eventually impose tighter AI constraints
“eventually, maybe the kind of US government. would agree on some kind of legislation that would set some more specific or tighter constraints around what isn't permissible.”
Listen at 3:43
Matthew Bloxhamon AI industry guidelinesPositive4:13
AI industry guidelines are preferable to legislation for now
“I think for now, industry guidelines would be the best way forward.”
Listen at 4:13
Matthew Bloxhamon Anthropic IPOPositive4:40
Anthropic likely remains interested in pursuing an IPO this year
“I would imagine they're still keen to go this year.”
Listen at 4:40
Matthew Bloxhamon Anthropic IPOPositive4:40
Anthropic likely remains interested in pursuing an IPO this year
“I would imagine they're still keen to go this year. That's just my view, though.”
Listen at 4:40
Matthew Bloxhamon IPO schedulesNeutral5:01
IPO timing is highly sensitive to stock-market sentiment
“the IPO schedules are incredibly sensitive to stock market sentiment.”
Listen at 5:01
Matthew Bloxhamon AI demandPositive5:39
AI demand will not slow in the near term
“I don't see any... near-term slowdown in demand”
Listen at 5:39
Matthew Bloxhamon AI demandPositive5:59
AI demand outlook will remain robust
“the demand outlook's still going to look pretty robust.”
Listen at 5:59
Matthew Bloxhamon AI investmentsPositive6:37
AI investment will shift toward improving existing models and differentiation
“a lot of that. investment is going to find its way into other parts of their existing models that they can make more competitive to kind of create differentiation”
Listen at 6:37
Matthew Bloxhamon AI model efficiencyPositive6:57
AI investment may target model efficiency, power consumption, and speed
“they start to look at the boundaries around efficiency of these models in terms of power consumption. or speed or other parameters.”
Listen at 6:57
Matthew Bloxhamon AI industry frameworkPositive7:28
AI companies will pursue a voluntary industry framework soon
“the next or kind of near term. goal would be to create some kind of self-voluntary industry framework that they'll all agree to abide by.”
Listen at 7:28
Matthew Bloxhamon AI regulationNegative7:51
AI legislation is not expected to be imposed soon
“I don't think anybody was expecting any legislation to kind of be imposed anytime soon.”
Listen at 7:51
Matthew Bloxhamon U.S. AI regulationMixed8:00
Future US AI legislation will likely be relatively light-touch
“whatever legislation comes is going to be fairly light touch.”
Listen at 8:00
Matthew Bloxhamon European AI adoptionMixed8:38
European AI adoption is good but faces more checks and balances
“Adoption is pretty good, but I think perhaps that there's already more checks and balances in most European markets”
Listen at 8:38
Matthew Bloxhamon European AI adoptionPositive8:38
European AI adoption is currently strong.
“Adoption is pretty good”
Listen at 8:38

Statements are attributed to the speaker as said on the episode and reflect their view at the time, not PodLume's. They are not advice.

Weekly email

State of the Pod, every Monday

The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.