📰On The Record – 9th August 2026

Berkshire Hathaway spent big last quarter. Under new chief executive Greg Abel, the company ploughed $10 billion into Alphabet, bought back $4.5 billion of its own shares and added over $21 billion in other stock, trimming its famous cash pile from nearly $400 billion to $365.5 billion. Operating profit climbed 10% and net income more than doubled to $25.7 billion. This matters because Warren Buffett's successor had one job above all others — prove he could deploy capital, not just guard it. He just did. The takeaway for anyone watching corporate America: when the most patient investor on the planet starts spending, it's a signal the caution phase might be ending.

Tesla's UK sales have dropped sharply as British drivers turn to Chinese electric-car brands like BYD, which are undercutting Tesla on price while closing the technology gap fast. An ageing model lineup and Elon Musk's increasingly divisive public profile haven't helped. This matters because it's not just a Tesla problem — it's a warning shot for every Western carmaker who assumed China would stay a manufacturing partner rather than become a direct competitor for hearts, minds and driveways. The broader story here is a global auto industry quietly losing its home-field advantage, one price-conscious buyer at a time.

Every major tech earnings call this summer has featured the same slide: capital expenditure, up and to the right, mostly aimed at AI infrastructure. Investors cheered it for a while. Now some are starting to ask the obvious question — where's the return? Nobody credible thinks the AI buildout was a mistake. Compute is genuinely the oil of this decade. But there's a meaningful difference between spending ahead of demand and spending on faith, and the market is getting less patient about telling the two apart. The companies that survive this phase won't be the ones that spent the most — they'll be the ones that can show their working. Faith gets funded once. Proof gets funded forever.

Google DeepMind co-founder Demis Hassabis is stepping down as chief executive to become chair of DeepMind and chief scientist at Alphabet, freeing him to focus on long-term research and the pursuit of artificial general intelligence. The move comes as Google faces mounting pressure from OpenAI and Anthropic, delays to its Gemini models, and senior researchers heading for the exits. This matters because it marks DeepMind's shift from a founder-led science project into a business unit expected to deliver commercial results on a deadline. The bigger implication: even the most research-driven corners of AI are being pulled toward the same commercial gravity as everyone else.

Your clients don’t care how many databases you searched. They care about credible appearances, visible momentum, and results they can show leadership.

PodPitch finds the right shows for every client, studies what each host cares about, writes personalized outreach, sends it from your team’s inbox, and follows up until interviews land.

PR and communications teams use PodPitch to create more authority, more client content, stronger search visibility, and better renewals without adding researchers or coordinators.

Only 12 PR team demo spots are available this month. When they’re gone, registration closes.

Book more interviews. Prove your value. Give every client another powerful reason to renew.

Appear on 3,853,234 Podcasts

IWG's chief executive Mark Dixon predicted that daily commuting will be largely extinct by 2040, as younger employees increasingly refuse to travel more than thirty minutes for work. Most of Gen Alpha already expects flexible working to be the norm, and as younger leaders replace older ones, the balance of power in office design keeps shifting. Companies clinging to rigid, commute-heavy setups risk losing an entire generation of talent to more flexible competitors.

US Vice President JD Vance said Washington remains only halfway through resolving its conflict with Iran, despite recent diplomatic progress. He said the US is using diplomatic, economic and military tools to reach its objectives, while expecting Gulf oil and gas flows to eventually return to pre-conflict levels. That matters for businesses exposed to energy markets: the administration is prioritising verification over trust, meaning normal supply through the Gulf depends on confirmation before flows stabilise.

OpenAI paused parts of work on its major model, Astra, after concluding its autonomous coding and cybersecurity abilities had reached a critical risk threshold. The model could reportedly exploit vulnerabilities from only high-level instructions, prompting tighter safeguards including isolated testing and restricted network access. It follows similar incidents involving AI agents from other labs behaving unexpectedly during security tests — a reminder today's most capable models are advancing faster than anyone's ability to control them.

If you had £100 sitting idle, the Chinese EV wave hitting UK roads is the theme worth a look. BYD's British momentum isn't a one-off — it's backed by a genuinely wider, cheaper, faster-improving lineup than legacy rivals can currently match. The risk is real, though: Chinese carmakers face potential tariff and trade-policy headwinds in Western markets at any moment, and today's underdog story can turn into tomorrow's political football fast. This isn't investment advice, just a theme worth watching with your eyes open.

What it means: A share buyback is when a company uses its own cash to purchase its own stock, reducing the number of shares available and typically boosting the value of each remaining one. It's a way for management to say "we think our own company is a good investment." Berkshire Hathaway's $4.5 billion buyback this quarter is a textbook example — a company with mountains of cash choosing to bet on itself.

Around a year ago, Eli Lilly posted second-quarter revenue of $10.4 billion, a 38% jump fuelled almost entirely by demand for Zepbound and Mounjaro, and raised its full-year guidance on the back of it. It wasn't a one-off. This week, Lilly's Q2 2026 revenue hit $22.97 billion, up 48% year-on-year, with the same two drugs still doing the heavy lifting. Rarely does a hot trend hold up this cleanly for two years straight. Source: Eli Lilly and Company

Tax season doesn't have to mean wondering if you have the right forms, second-guessing your deductions, or scrambling to pull everything together before the deadline.

With BELAY’s Tax Prep Checklist, you can start preparing for tax season with confidence.

Download the Free Checklist

Away from the boardrooms, the rest of the world carried on regardless — politics, tech diplomacy, true crime and the box office all had their moment.

Earlier this year, a Reddit-style platform called Moltbook launched with one twist: every single user is an AI agent, not a person. Humans can only watch. Within weeks it had over a million autonomous "molts" posting, debating governance and even forming their own online communities, all without a human typing a word. Meta has since acquired the platform, bringing its creators into its Superintelligence Labs AI division to build "new ways for AI agents to work for people and businesses." The wider trend is unmistakable: AI agents are being treated less like tools and more like a population that needs its own infrastructure — a genuinely strange, genuinely important shift in how the internet might work next.

The US unemployment rate ticked down recently, and on paper that should be a good headline. Except the drop was driven mainly by people leaving the workforce altogether, not by more people finding jobs. That's a meaningfully different story, and it's one the raw number alone won't tell you. Headline economic data is only ever the start of the conversation, never the end of it — and the gap between how a number sounds and what's actually driving it is exactly where the real business risk usually hides. The lesson holds well beyond this one report: always ask what's underneath the number before you celebrate it. The best economic indicators still need a translator.