AI News Weekly Podcast #1

Welcome to the AI News Weekly Podcast from AI News dot Today. This is episode 1.

This week, Anthropic emerged as SpaceX's biggest customer, Reddit said it will become a mostly logged-in site as search traffic falls, and South Korea proposed a record budget that nearly doubles chip and AI spending. We also cover Australia's annual cyber threat report and OpenAI's policy push after the US midterms.

Australia's cyber agency urges AI defenses as incidents rise 20%

The Australian Signals Directorate, the country's signals intelligence agency, has urged businesses and critical infrastructure operators to adopt advanced AI models for cyber defense. In its annual cyber threat report, released on October 27, the agency said it responded to 1,500 incidents in the 2025-26 year, up 20 percent. It notified organizations of more than 4,400 potential malicious incidents, an increase of 80 percent. The report said hostile foreign governments, criminals and hackers are using AI to find vulnerabilities and launch attacks faster than before. The agency also told Australians to expect more important online services to go offline unexpectedly, saying system security must be prioritised over availability. Defence Minister Richard Marles said Australians must enhance their cyber defenses while harnessing AI's benefit.

Reddit plans logged-in experience as search referrals fall

Reddit says it is moving to a primarily logged-in experience, a change announced on its third-quarter earnings call as traffic from search engines declines. The company said search referrals fell by more than 1 million a week on average in October, as AI-generated summaries answer more queries without a site visit. Chief executive Steve Huffman told Axios the change is meant to improve the user experience, strengthen the ads business and stop bots from scraping Reddit's data. TechCrunch reported that protecting that data also positions Reddit to sell access to model developers. US daily users fell to 50.7 million from 53.2 million in the second quarter, even as revenue rose 53 percent to 895 million dollars. Testing begins in select markets outside the US in the fourth quarter.

Anthropic becomes SpaceX’s biggest customer with a third of sales

Anthropic became SpaceX's biggest customer last quarter, accounting for a third of the company's sales, Semafor reported on October 30, citing SpaceX's quarterly results. According to Semafor, the figure feeds a growing worry among investors that the AI industry and the wider economy depend on a small number of companies, raising the risk of market instability if one of them fails. The dependence runs in more than one direction. Semafor reported that Anthropic's IPO prospectus, made public in the week of October 26, shows almost half of its revenue comes from Amazon, Google and Microsoft. Bridgewater warned clients that such concentration means more plain luck is needed to keep economies going. SpaceX, meanwhile, has closed 40 billion dollars of debt to buy GPUs, Elon Musk wrote on X.

South Korea nearly doubles chip and AI spending in record budget

In Seoul, President Lee Jae Myung proposed a record budget of 820.9 trillion won for 2027, about 610 billion dollars. It sets aside 21.3 trillion won for semiconductors, physical AI and AI data centers, nearly double the 2026 level. Physical AI refers to systems that run robots and other machines in the real world. A separate 4.7 trillion won will go toward a sovereign AI model, one built and controlled at home. Total spending would rise 12.8 percent, and Lee said there would be virtually no net increase in government debt. The plan ties more state money to the industry that has driven much of the country's growth. Exports rose 93.7 percent from a year earlier to a record 115.3 billion dollars in October.

OpenAI to rebrand data centers as 'AI compute campuses'

OpenAI plans a policy and public relations push right after the November 3 US midterm elections, including an effort to rebrand data centers as AI compute campuses, Axios reported. The campaign is aimed at a Washington that Axios expects to be harsher on AI companies, with Democrats likely to win the House. Chris Lehane, OpenAI's chief global affairs officer, said his team has been stockpiling policy proposals since August, among them AI speed limits and a US role in developing global standards. He called the public argument over data centers good and healthy, saying communities should hold companies accountable. The push begins in the lame-duck session of Congress, which opens on November 9. Separately, the Financial Times reported that OpenAI cut its fundraising target by 10 billion dollars, to about 20 billion.

That's the week. The full stories are at AI News dot Today, and you can subscribe to the AI News Weekly Podcast wherever you get your podcasts.