Yesterday, chip stocks had their worst session in months. Korea's market dropped 10%, dragging Samsung, SK Hynix, and then every U.S. memory name down with it.

Micron fell 13%. SanDisk fell 11%.

Tonight, Micron reports earnings. If demand for AI memory chips is still strong, yesterday was noise. If it's not, the companies that built hiring plans around AI infrastructure spending are going to start rethinking them.

I'll break down what triggered the selloff, what Korea's leveraged ETF market had to do with it, and what to watch in tonight's numbers. Plus: oil dropped below $80 on Iran deal progress, consumer sentiment inched up from its record low, and BCG found that the workers getting the most from AI aren't the ones with the best tools.

Let's get into it.

Oil dropped below $80 for the first time since March after the US and Iran signed a framework to end their conflict and reopen the Strait of Hormuz. Brent crude fell 5% to $78.96 and WTI settled at $76.05. Analysts say gas prices won't return to pre-war levels until 2027 even if the ceasefire holds, because global oil inventories need months to recover.

Consumer sentiment rebounded 9% to 48.9 in early June, up from May's all-time low of 44.8. The improvement showed up across age, education, and political groups, with lower-income consumers leading the bounce as gas prices eased. Still 19% below a year ago, and year-ahead inflation expectations are holding at 4.6%.

Meta is investing $900 million into India's CRED at a $4.5 billion valuation, taking a 20% stake in the fintech platform that handles 40% of India's credit card bill payments. The deal comes with a leadership swap: CRED founder Kunal Shah is leaving to run WhatsApp globally, replacing Will Cathcart.

NVIDIA launched the BioNeMo Agent Toolkit at the BIO International Convention, giving AI agents tools for protein design, drug discovery, and genomic analysis. More than 50 companies including Lilly and Schrodinger are already using it. Both Anthropic and OpenAI are integrating the toolkit into their platforms.

GitLab cut 14% of its workforce and exited 22 countries as part of what CEO Bill Staples called the company's pivot to the "agentic era." Revenue was up 23% year-over-year when the cuts happened. GitLab is reinvesting most of the $30-35 million in savings back into R&D and its Duo Agent Platform.

Monday 7am. Three messages in #growth.

Stripe revenue by channel, Meta and Google spend reconciled against GA4, Klaviyo flow performance, Shopify AOV by source. Posted by Viktor at 6am.

The campaign brief he wrote sits in #campaigns. Brand monitoring scrape runs every six hours. Competitor pricing update lands every Friday.

Your media buyer, content lead, and CMO open Slack to the same prepared room. 3,000+ integrations including every ad platform, CDP, and CMS you run.

"Viktor is like the most capable all-round colleague you can imagine." Sam, CEO, Givr.

Get Started for Free

Yesterday (June 23), South Korea's Kospi index dropped 9.99%, its steepest fall in more than three months, triggering circuit breakers twice during the session. SK Hynix fell more than 11%. Samsung slid over 8%. By afternoon, the selloff had crossed the Pacific: Micron dropped 13% to close at $1,051, SanDisk plunged 11%, and Western Digital took a similar hit.

What made it this bad? A few things stacked up at once. Korean regulators had recently approved 16 leveraged ETFs tracking Samsung and SK Hynix, and when profit-taking started, the leverage amplified the move. On top of that, Micron insiders had sold $92.5 million in shares ahead of tonight's earnings, which didn't exactly build confidence.

The broader setup was already fragile. Micron trades at nearly 50x earnings, well above its five-year median of around 21x. Buy-side expectations are running ahead of the company's own guidance. SanDisk was up 858% year-to-date before the crash. When a trade gets that crowded on a single thesis, any crack in confidence gets punished fast.

The real question isn't whether the numbers were overextended. They were. It's whether this is a valuation correction or an early signal that AI infrastructure spending is slowing down. These stocks ran on one thesis: cloud providers and AI companies need enormous amounts of high-bandwidth memory to train and run models. That thesis either holds or it doesn't.

Why it matters: Micron reports fiscal Q3 earnings tonight (June 24) after market close. That report will either confirm that HBM demand is holding, or it won't. If the numbers disappoint, the "AI spending slowdown" narrative accelerates, and companies that built headcount around AI infrastructure expansion may start hedging their plans. If Micron beats, yesterday was just Korean leverage unwinding and the thesis stays intact.

What to do: If you're at a company whose recent hiring or expansion was connected to AI infrastructure spending, tonight's report is worth watching. Pull up your company's latest earnings call transcript and search for "AI," "infrastructure," or "capex." If those words show up more than five times in connection with growth plans, you have some exposure to how this plays out.

BCG surveyed nearly 12,000 employees across more than a dozen markets and found something that flips the typical AI conversation. Adoption is no longer the bottleneck. 74% of frontline employees are now regular AI users, up 23 percentage points year-over-year. The 42% who use it most report saving roughly a full workday per week.

But here's the disconnect. Two-thirds of those workers say they get limited or no guidance on what to do with the time they save. More than half aren't reinvesting that time into more strategic work. They're just faster at the same tasks.

The standout finding: companies with a clear AI strategy see 25 percentage points more business impact than companies that just rolled out better tools. Better tools alone moved the needle by about 5 points. Strategy outperformed tooling 5-to-1.

There's also a tension BCG calls the "joy paradox." 67% of regular AI users say it has improved their job satisfaction. But 41% also report increased cognitive load. AI is making work better and harder at the same time.

Why it matters: If your organization handed everyone Copilot or Claude and called it an AI strategy, this data suggests you're leaving most of the value on the table. The edge isn't access to AI. It's knowing what to do with it.

→ Land your next role faster with Kimchi, our Claude-powered job search tool: Start searching

→ If you're targeting $150K+ roles, have our team run your entire job search for you: Apply here

→ Build your own personal AI workforce handling your work, business, and personal life: Access to ready-made agents

Pull up your company's most recent earnings call transcript. Search "[company name] Q1 2026 earnings transcript" on Google. Hit Ctrl+F and search for "AI," "infrastructure," and "capex." If those terms appear more than five times in connection with growth, hiring, or expansion plans, your company has meaningful exposure to the AI infrastructure cycle. Then check the investor Q&A section for analyst questions about AI spending sustainability. If analysts are already asking, the market is already watching.

Think about it. The generation that's supposed to fill Director and VP roles next.

The answer: 6%.

That's from Deloitte's 2026 Global Gen Z and Millennial Survey. The top reasons: fear of burnout, stress, and damage to work-life balance. For anyone currently in a senior role, this number explains why your succession pipeline feels thinner than it should.