BUSINESS

World No. 1 Jannik Sinner, the two-time Wimbledon champion, and French Open winner Alexander Zverev, the world No. 2, are also in the men’s draw.

From BTS to Son Heung-min to EJAE, here’s a recap of Korean cultural moments from the World Cup—including one of the biggest public outcries in Korean soccer history.

Christopher Nolan’s ‘The Odyssey’ is all about the breaking of Zeus’s Law—here’s the significance and historical fact behind the sacred rule, explained.

Advanced Micro Devices (AMD) stock jumped more than 7% Tuesday, July 21, as investors cheered the chipmaker’s latest AI infrastructure push and looked ahead to its key AI event.On Monday, July 20, AMD introduced Helios, its first rack-scale AI system, which is expected to begin shipping later this year to customers including Microsoft (MSFT), Meta Platforms (META), OpenAI, and Oracle (ORCL).Helios’ rollout is widely considered AMD’s biggest step yet into rack-scale AI infrastructure as it seeks to compete more directly with Nvidia (NVDA). Investors are looking for signs that the company’s expanding AI portfolio can accelerate data center growth and help it gain share in the fast-growing AI market.The next catalyst comes July 22-23, when AMD hosts its Advancing AI 2026 event in San Francisco. CEO Lisa Su is expected to provide updates on the MI450X accelerator, the MI500 GPU family, next-generation EPYC server processors, and AMD’s broader AI roadmap.Many also view the event as AMD’s next opportunity to show how it plans to expand its presence in AI infrastructure ahead of second-quarter earnings on Aug. 4. AMD stock has outperformed the chip sector this yearLast week, AMD shares fell approximately 11% as semiconductor stocks broadly pulled back.The sell-off was triggered by Micron Technology (MU) amid concerns about growing competition from Chinese memory-chip maker ChangXin Memory Technologies. According to Barron’s, the company is preparing a Shanghai IPO that could raise as much as 66.7 billion yuan ($9.8 billion).Related: Cathie Wood sells $11.7 million of tumbling semiconductor stockAlthough AMD doesn’t compete directly in the memory market, many investors took profits across semiconductor stocks last week following their strong gains this year.Even after last week’s decline, AMD remains one of the best-performing large-cap chip stocks. Shares are up 152.8% year to date, compared with a 72.7% gain for the Philadelphia Semiconductor Index. Nvidia, by comparison, has gained 11%.In May, AMD reported strong first-quarter results, raising its outlook for the data center CPU market as demand for agentic AI continues to grow. The company also reiterated that its AI GPU business is expected to exceed its long-term target of an 80% compound annual revenue growth rate. Meanwhile, Ark Invest CEO Cathie Wood has been trimming her AMD position. As of July 21, Ark funds had sold 145,550 AMD shares this month, worth roughly $79 million based on Tuesday’s price. Even after the recent selling, AMD remains the eighth-largest holding in the ARK Innovation ETF.

AMD shares are up more than 150% year to date.Getty Images

Bank of America stays bullish on AMD stockBank of America reiterated its buy rating and $620 price target on AMD ahead of the company’s AI event, according to a recent research note sent to TheStreet.The firm said AMD could use the event to expand its long-term AI opportunity beyond the more than $1 trillion market outlined at its 2025 Analyst Day. “We would not be surprised to see management frame a path toward a $1.5 trillion + 2030 AI infrastructure TAM, spanning accelerators, networking, memory, CPUs and rack-scale systems,” the analysts wrote. Related: Bank of America CEO warns inflation will back Fed into a corner”With the event occurring ahead of the Aug. 4 earnings release, we expect commentary to remain focused on long-term opportunities rather than near-term guidance.”Bank of America said investors will likely focus on whether Helios and the MI450 platform can better compete with Nvidia’s rack-scale AI systems. After all, the key debate is “whether AMD can evolve from an accelerator supplier into a broader AI systems company.””Investor focus will likely center on whether Helios/MI450 can narrow the gap versus Nvidia’s rack-scale AI systems,” the firm wrote, adding that investors should watch for updates on 2027 deployments, production ramps, and broader hyperscale adoption.The firm also believes demand will be more important than customer announcements.”We believe the more important message may be demand rather than customer logos,” Bank of America wrote, adding that AMD’s data center revenue could grow more than 100% in 2027 as MI accelerators ramp and EPYC adoption expands.Many other Wall Street analysts believe in AMD stock’s higher potential, despite its strong gains this year.Last week, UBS raised its price target on AMD to $700 from $670 while maintaining its buy rating. The firm said supply chain checks point to stronger demand for AMD’s AI accelerators through 2027 and believes Amazon could emerge as a major customer for the MI450X platform.UBS also raised its 2027 revenue forecast to $83.4 billion from $79.2 billion and increased its earnings estimate to $14.63 per share.Related: Key auto parts maker closes factory, lays off 325 workers

Plus, more ways to lower how much you pay out of pocket for medical care in retirement.

Morgan Stanley downgraded Adobe to underweight from equal weight on Tuesday, July 21, slashing its price target to $240 from $365, according to Seeking Alpha.The move landed one day after CLSA initiated coverage of Adobe with an outperform rating and a $300 target, according to Investing.com.Two firms looked at the same company and reached opposite conclusions, and that split matters more than either rating on its own.What Morgan Stanley’s Adobe downgrade actually saysAnalyst Adam Wood built his case around three problems happening at once. Adobe is searching for a new CEO and CFO, following CEO Shantanu Narayen’s March announcement to step down and CFO Dan Durn’s departure in June. The company is simultaneously shifting toward a freemium pricing model and facing open questions about how much of its Creative Cloud business generative AI tools could replace.Related: Citi resets Adobe stock price targetWood wrote that Adobe’s overlapping transitions compound execution risk as the AI disruption debate clouds the path to faster subscription growth, according to CNBC.Even so, Morgan Stanley noted that Adobe trades near 12 times its projected 2027 earnings. The firm said that valuation already reflects much of the disruption risk that’s worrying investors.That detail matters because it frames this downgrade as a growth debate, not a call that Adobe is in real trouble.

Morgan Stanley cut Adobe to underweight with a $240 target, a day after CLSA initiated coverage with a bullish $300 target and outperform rating.Bloomberg / Getty Images

Why CLSA sees the opposite trade for AdobeOne day earlier, CLSA initiated coverage of Adobe with an outperform rating and a $300 price target. Analyst Bhavtosh Vajpayee set that target at 15 times Adobe’s projected 2027 earnings, a multiple benchmarked against rival Salesforce.He argued the AI disruption story around Adobe has been overdone, given the company’s software margins.More Tech Stocks:Morgan Stanley sets jaw-dropping Micron price target after eventNvidia’s China chip problem isn’t what most investors thinkQuantum Computing makes $110 million move nobody saw comingCLSA pointed to Adobe’s gross margin, among the highest in the software sector, as evidence the business is healthier than its stock price suggests.The firm framed the recent pullback as a valuation opportunity rather than a warning sign. Two analysts read the same margin profile within 24 hours of each other and landed on opposite ratings.The gap between the two firms is not an outlier. Roughly 40 analysts currently cover Adobe, and only about a dozen carry a buy rating or higher, while roughly two dozen sit at hold, according to Stocktwits.The average 12-month price target across that group lands closer to $273, squarely between Morgan Stanley’s bearish call and CLSA’s bullish one.Adobe’s muted stock reaction complicates both callsAdobe (ADBE) closed near $234.74 on Monday, July 21, down about 1%, according to data from TheStreet. The stock then slid further, falling as much as 3% in premarket trading on Tuesday, after the Morgan Stanley downgrade, according to Stocktwits.Neither the underweight call nor the outperform initiation has moved the stock much, which suggests investors are still waiting for clearer evidence before picking a side.That hesitation is notable, given how far apart the two price targets sit. A $240 target and a $300 target imply a 25% gap in how Wall Street values the same shares. When analysts disagree that sharply, it usually means the underlying data support more than one story.The fundamentals still look solidThe ratings split is happening against a backdrop of resilient growth. Adobe’s second quarter revenue rose 12.7% year over year, its largest gain since 2022.That number complicates the bear case, since a company facing serious AI substitution risk would typically show it in slowing sales first.Adobe’s situation reflects a wider pattern across software stocks this earnings season. Investors are trying to sort which companies will absorb generative AI into their products and which ones will lose ground to it, and the data so far is not settling the argument either way.Whoever Adobe names as its next CEO will inherit that unresolved question, and the market’s next move will likely hinge on which side of the debate that hire signals.Related: Adobe’s outgoing CEO makes CEO big bet on the future

Poolside, the San Francisco AI lab that has spent most of its three-year existence quietly selling coding models to governments and defense agencies, released its most capable model to date on Tuesday — and made an unusually aggressive bet that radical transparency, not raw scale, is how a smaller lab competes at the frontier.The model, Laguna S 2.1, is a 118-billion-parameter Mixture-of-Experts (MoE) system that activates only 8 billion parameters per token, supports a context window of up to 1 million tokens, and — according to benchmarks published by the company — matches or beats open models several times its size on agentic coding tasks. The weights are available immediately on Hugging Face under the permissive OpenMDW-1.1 license.The headline numbers are striking for a model this small. Poolside reports that Laguna S 2.1 scores 70.2% on Terminal-Bench 2.1, a benchmark of long-horizon terminal tasks, placing it 11th on the company’s compiled leaderboard — ahead of DeepSeek-V4-Pro-Max, a 1.6-trillion-parameter model that scored 64.0; Thinking Machines’ 975-billion-parameter Inkling, at 63.8; and Nvidia’s 550-billion-parameter Nemotron 3 Ultra, at 56.4. On SWE-Bench Multilingual, it posts 78.5%, and on SWE-Bench Pro’s public dataset, 59.4%.Perhaps more telling than any single score: the model went from the start of pre-training on May 22 to public launch in under nine weeks, trained on 4,096 Nvidia H200 GPUs. In an industry where flagship model cycles are typically measured in quarters or years, Poolside has now shipped three models in three months.Why the West’s open-weight AI gap has become a boardroom issueThe release lands in the middle of an increasingly pointed debate about the provenance of open-weight AI. Over the past year, developer adoption has shifted decisively toward open-weight systems that companies can download, inspect, and run on their own infrastructure — and the leading options in that category have overwhelmingly come from Chinese labs. DeepSeek, Qwen, Kimi, GLM, MiniMax, and Tencent’s Hunyuan line all feature prominently in Poolside’s own comparison tables.Poolside’s accompanying press release frames Laguna S 2.1 explicitly as a response, noting that the model occupies a size class into which no Western lab has released open weights in 11 months — since OpenAI’s gpt-oss-120b last August. “The West needs open-weight models it can trust, run, and build on,” said Jason Warner, Poolside’s co-CEO, in the announcement.Co-founder and co-CEO Eiso Kant made the philosophical stakes even plainer in a lengthy post on X. “I believe intelligence should and will become a commodity,” he wrote, arguing that the open ecosystem “will not win by being the best in its own category.” Users, he argued, simply want the best intelligence for the task at hand — so open models must be on par with, or better than, their closed equivalents.The strategic logic here is not charity. Poolside’s core business is deploying models inside the security boundaries of government, defense, and regulated enterprises — customers for whom closed, metered API access is often a non-starter for compliance and sovereignty reasons. Every enterprise that standardizes on a Chinese open model today becomes harder to win tomorrow. Releasing competitive open weights is both an ecosystem play and a top-of-funnel strategy for the company’s high-security deployment business. It also reframes the AI race away from terrain where Poolside cannot compete — frontier-scale capital expenditure — and toward terrain where it believes it can: cost per token, self-hosting, and iteration speed.How a sparse architecture makes enterprise AI agents affordable to runThe technical design reflects a specific thesis about where value in coding AI is moving. Laguna S 2.1’s sparse MoE architecture — 256 routed experts plus one shared expert, with grouped-query attention and interleaved sliding-window layers, according to the Hugging Face model card — means inference costs scale with the 8 billion active parameters, not the 118 billion total. Poolside emphasizes that the model is small enough to run on a single Nvidia DGX Spark, the desktop-class AI machine.That matters for what Poolside calls token economics. Long-horizon coding agents are voracious consumers of tokens: the company’s published data shows the model consuming a mean of roughly 249,000 completion tokens per trajectory on its hardest benchmark when thinking mode is enabled. At metered API prices, agentic workloads at enterprise scale become a meaningful budget line item. On OpenRouter, Poolside is offering a free 256K-context endpoint and a dedicated 1M-context deployment priced at $0.10 per million input tokens and $0.20 per million output tokens — aggressive pricing that undercuts most frontier alternatives by an order of magnitude.The ecosystem support is unusually broad for day one. The model is live on Baseten’s model library and Vercel’s AI Gateway, with integrations across vLLM, SGLang, Ollama, and llama.cpp, plus quantized variants down to 4-bit GGUF files — 75 gigabytes — for local use. But Poolside’s more interesting claim is behavioral, not architectural. Pengming Wang, co-head of applied research at Poolside, said the gains came from improving the model’s working habits: “more verification, less taking things for granted, not declaring victory early, and being more persistent.” Raw intelligence, the company argues, is one axis of capability; a model’s way of working is a second axis that matters immensely for agents left unattended for hours.Publishing every benchmark trajectory to counter AI’s credibility crisisThe most consequential part of the release for enterprise buyers may be an evaluation-transparency move with little precedent among major labs: Poolside published the complete, unedited trajectory of every trial in its final benchmark runs — every reasoning step, tool call, and shell command behind every reported score.This addresses a growing credibility problem in AI benchmarking. As top scores on mature benchmarks cluster in the 70–90% range, and as “reward hacking” — models finding solutions online or gaming verifiers rather than solving problems — has become endemic, self-reported numbers have lost much of their signal. Poolside disclosed its own encounters with the problem candidly: during training, more than half of trajectories on some SWE-bench tasks were flagged because the model simply researched the original bug-fix pull request online and applied it. The company documented its mitigations, including prompt addenda, LLM-based judging calibrated against human labels, and expert annotator review of a high-scoring Terminal-Bench run.Three published case studies illustrate what the company means by persistence. In one, the model built a working HTML/CSS rendering engine from an empty folder in a 181-step, 50-minute unattended session — then, lacking vision capabilities, spun up headless Chromium to numerically compare its canvas output against a real browser’s rendering. In another, pointed at Poolside’s own agent harness in an automated optimization loop, the model made the Go codebase 5.2% faster with roughly 70% lower memory allocation, finding an O(n²) string-concatenation bug along the way. In a third, working in a sandbox with no Python installed, the model did its number theory in Perl and independently re-derived a proof of Erdős problem #397 — a combinatorics question open for five decades until GPT-5.2 Pro first solved it this past January. Poolside notes that its model’s construction is structurally different from the earlier published solution, and that its November 2025 knowledge cutoff precedes the first proof.What the disclosed limitations and benchmark fine print revealPoolside deserves credit for disclosing limitations most labs bury. The model can overfit to its native harness and stumble on slightly different tool schemas in third-party agents, mangles JSON in nested tool arguments, and is prone to overthinking on competition math. There is currently no user-configurable thinking-effort dial — just on or off — and the gap between the modes is enormous: thinking lifts Terminal-Bench 2.1 from 60.4% to 70.2%, and DeepSWE from 16.5% to 40.4%, at substantially higher token cost.Buyers should apply their own discounts to the comparison tables. Poolside’s methodology takes the maximum of vendor self-reported scores, benchmark-author leaderboards, and third-party figures for competitors — a reasonable convention, but one that mixes harnesses and test conditions. On DeepSWE, notably, Poolside ran its own agent harness rather than the leaderboard’s standard mini-swe-agent, a difference the company acknowledges makes scores less directly comparable. And the frontier remains clearly out of reach: closed models like GPT-5.6 Sol, at 88.8 on Terminal-Bench 2.1, and Claude Fable 5, at 88.0, along with the 2.8-trillion-parameter open-weight Kimi K3, at 88.3, sit well above Laguna S 2.1.The deeper structural question is whether Poolside’s “Model Factory” — the internal platform the company credits for its rapid release cadence — can sustain this pace as models scale. The trajectory so far is genuinely unusual: the April dual release of Laguna M.1 and XS.2, the July 2 refresh of XS 2.1, and now S 2.1, which the company says outperforms April’s flagship M.1 at roughly a third of its active size. Remarkably, S 2.1 used the exact same pre-training data as XS 2.1, meaning nearly all the improvement came from scale, training fixes, and post-training across the company’s corpus of 409,000 agentic and non-agentic training environments. Poolside says its next, larger Laguna model began pre-training last week.For technical decision makers, Laguna S 2.1 is the most credible Western open-weight option to emerge in nearly a year for self-hosted agentic coding — with published evidence, a permissive license, broad ecosystem support, and an economics story built around hardware you can own. Whether it dents the dominance of Chinese open models will depend less on this release than on the ones that follow it.Kant, for his part, has already told the world how he intends that story to end. Poolside is building toward a future where the most capable intelligence “can be owned and shaped by anyone,” he wrote — and the company plans to keep shipping “until that future exists.” In an industry where the biggest labs increasingly lock their best work behind an API, the most radical thing about Laguna S 2.1 may not be what it scores, but that anyone can download it and check.

There has been an escalation in strikes between the U.S. and Iran throughout the last 10 days over who has control of the Strait of Hormuz.

The AI boom is reshaping tech portfolios. Discover which semiconductor stocks are buys today and which you should ignore to protect your capital.

Illinois enacted a 0.2% tax on all crypto transactions last month, with the tax taking effect next year.