The Conversation (0)
Tech 5: TSMC, ASML Release Latest Results, NVIDIA to Resume Sales to China
“The difference between the IT bubble in the 1990s and the AI bubble today is that the top 10 companies in the S&P 500 (INDEXSP:.INX) today are more overvalued than they were in the 1990s,” he wrote on Wednesday (July 16).
Moor Insights & Strategy founder Patrick Moorhead expressed similar thoughts last week.
However, Sanctuary Wealth's chief investment strategist, Mary Ann Bartels, told CNBC’s Power Lunch team that valuations are justified by the technology that’s being unleashed. Major financial firms like Citigroup (NYSE:C), Bank of America (NYSE:BAC) and Morgan Stanley (NYSE:MS) also said they are increasingly exploring digital asset offerings, signaling traditional finance's growing involvement in crypto and the broader adoption of innovative technologies.
These announcements came alongside positive earnings reports and mixed inflation data that helped lift markets to renewed highs, culminating in global manufacturer 3M (NYSE:MMM) raising its full-year profit forecast on Friday.
The company is projecting a smaller tariff-related hit to its 2025 earnings.
1. TSMC, ASML release latest quarterly results
This week saw semiconductor giants Taiwan Semiconductor Manufacturing Company (TSMC) (NYSE:TSM) and ASML Holding (NASDAQ:ASML) report their latest quarterly earnings.
The companies received vastly different reactions from the market. Contract chipmaker TSMC saw its valuation soar on Thursday (July 17) morning after it posted record profits that exceeded expectations and raised its full-year revenue forecast by 30 percent due to demand for artificial intelligence (AI) chips.
While the chipmaker addressed minor concerns about US tariffs and inventory, AI-driven growth dominated investor sentiment. Shares of TSMC opened 4.51 percent higher from Wednesday’s (July 16) closing price.
Positive sentiment spilled over into other chip stocks, with NVIDIA (NASDAQ:NVDA) and Broadcom (NASDAQ:AVGO) also seeing gains. TSMC maintained its position to close up 5.87 percent for the week.
TSMC and ASML performance, July 15 to 18, 2025.
Chart via Google Finance.
Conversely, ASML, a lithography systems monopolist, saw its share price plunge more than 8 percent ahead of Wednesday’s open, despite solid Q2 numbers, due to a cautious outlook for late 2025 and 2026.
In a statement, the company said it cannot confirm growth in 2026 due to current macroeconomic and geopolitical developments. ASML closed the week 7.39 percent below its Monday opening price.
The divergence highlights their supply chain positions: TSMC directly benefits from the immediate AI boom, while the prospects for ASML, a step removed, remain uncertain.
2. US announces major investments in Pennsylvania
US President Donald Trump joined Pennsylvania Senator Dave McCormick (R) at the inaugural Energy and Innovation Summit at Carnegie Mellon University in Pittsburgh on Tuesday (July 15).
He announced an investment amounting to over US$90 billion in AI and energy infrastructure in the state.
The sum outlined by Trump covers several multibillion-dollar spending plans from the likes of Google (NASDAQ:GOOGL), Blackstone (NYSE:BX), Anthropic, GE Verona (NYSE:GEV) and others for power generation and grid modernization. It also includes natural gas production to help power data centers.
Additionally, the preview mentions AI training programs and apprenticeships for businesses.
“These commitments will create tens of thousands of construction jobs and thousands of permanent jobs, signaling Pennsylvania’s readiness to power the AI and energy revolution, further strengthening America’s resilience and independence,” McCormick’s office wrote in a press release.
Separately, Google and Brookfield Asset Management (NYSE:BAM) announced on Tuesday that they have entered into a framework agreement to provide up to 3,000 MW megawatts of domestically produced hydropower from Brookfield's Holtwood and Safe Harbor hydroelectric facilities in Pennsylvania. The agreement allows for future expansion, with an initial focus on the mid-Atlantic and mid-continent electricity markets.
3. NVIDIA to resume chip sales to China
On Monday, NVIDIA CEO Jensen Huang said his company will resume H20 GPUs sales to China after productive meetings with government officials from the US and Beijing earlier this month.
In a press release, the company said it has been assured by the US government that licenses will be granted.
NVIDIA performance, July 15 to 18, 2025.
Chart via Google Finance.
Shares of the chipmaker opened 4.27 percent higher on Tuesday and closed the week up 4.25 percent.
4. Apple to invest in US rare earths miner
On Tuesday, Apple (NASDAQ:AAPL) said it will invest US$500 million in rare earths miner MP Materials (NYSE:MP) as part of an effort to strengthen the American rare earths supply chain.
MP is the only fully integrated rare earths miner operating in the US. Last week, the US Department of Defense said it would buy a direct equity stake in the company, becoming its largest shareholder.
The company's Apple collaboration includes plans to build out MP’s neodymium magnet manufacturing lines at its Texas factory specifically for Apple products. This expansion is slated to boost production and create jobs in advanced manufacturing and research and development, helping to meet global demand.
Apple and MP will also collaborate to establish a rare earths recycling line in Mountain Pass, California, and will develop new magnet materials and processing technologies to improve magnet performance.
“American innovation drives everything we do at Apple, and we’re proud to deepen our investment in the U.S. economy,” said Tim Cook, Apple’s CEO.
5. OpenAI and AWS launch new AI agent features
Open AI has launched a powerful new Agent mode in ChatGPT for pro, plus and team users.
It can autonomously complete tasks across the web, and also includes productivity tools.
The new feature enables AI agents that can help automate workflow by creating and editing spreadsheets and presentations, generating reports, analyzing data and managing calendars on users’ desktops; agents can also browse websites and fill out forms with user approval. The company has plans to add e-commerce checkouts.
Aside from that, the Financial Times reported this week that OpenAI plans to take a cut of online shopping purchases made within its chatbot as a way to generate revenue from people using AI for shopping inspiration.
Amazon (NASDAQ:AMZN) also made major announcements around AI agents this week. At its Amazon Web Services (AWS) Summit in New York, the company launched Bedrock AgentCore, a suite of enterprise-grade services that will allow developers to build, deploy and run scalable agents. AWS also introduced AI Agents & Tools, a new category on AWS Marketplace. It features pre-built agents from partners like Anthropic, IBM (NYSE:IBM) and Stripe.
Don't forget to follow us @INN_Technology for real-time news updates!
Securities Disclosure: I, Meagen Seatter, hold no direct investment interest in any company mentioned in this article.
AI Market Update: Q2 2025 in Review
While the race for compute, energy, hardware and supply chain dominance intensified, talk of tariff policies reigniting inflation or contributing to stagflation created brief periods of contraction.
Concerns also grew around AI-driven job displacement, amplified by Anthropic CEO Dario Amodei’s ominous warning that AI could eliminate up to half of all entry-level white-collar jobs within the next five years.
On a more positive note, the S&P 500 (INDEXSP:.INX) and Nasdaq Composite (INDEXNASDAQ:.IXIC) both ended Q2 up by 0.5 percent, closing the first half of 2025 at all-time highs with gains of 5.5 percent.
That said, investor enthusiasm for AI is showing early signs of recalibration.
Big Tech delivered generally robust Q2 earnings despite initial volatility in April, but posted only modest year-to-date gains, suggesting near-term caution around richly valued growth names. Meanwhile, quantum computing, which NVIDIA (NASDAQ:NVDA) CEO Jensen Huang said was decades away just six months ago, made measurable progress in Q2, drawing attention from both deep-tech investors and national governments.
McKinsey’s annual Quantum Technology Monitor projects that quantum computing, communication and sensing could generate up to US$97 billion in global revenue by 2035, with quantum computing leading the way.
Not surprisingly, AI companies performed well. Thirty-eight AI stocks chosen by Morningstar — including Palantir Technologies (NASDAQ:PLTR), Palo Alto Networks (NASDAQ:PANW), Synopsys (NASDAQ:SNPS) and Micron Technology (NASDAQ:MU) — closed 27.3 percent higher, outpacing the Morningstar US Technology Index, which gained 22 percent.
Ultimately, the quarter underscored a strategic pivot for major tech players, prioritizing vast infrastructure investments alongside aggressive AI monetization efforts to capitalize on this transformative era.
AI results impact major tech players
In public markets, AI-related equities continued to attract attention.
NVIDIA posted another blockbuster quarter, with its market cap on the cusp of $US4 trillion at the end of June. Its performance was driven largely by demand for Blackwell architecture.
Alphabet (NASDAQ:GOOGL), facing a possible Chrome divestiture, reported an increase in AI-related ad revenue and highlighted growing adoption of its Gemini model suite. Amazon (NASDAQ:AMZN) reported a 23 percent annual increase in net sales from its Amazon Web Services segment, beating earnings estimates by 17.78 percent.
Meta Platforms' (NASDAQ:META) Reality Labs division reported a $US4.2 billion operating loss; however, interest in embodied AI applications for the metaverse and augmented reality continue to be the company’s long-term play, with CEO Mark Zuckerberg poaching the industry’s top talent to assemble the Meta Superintelligence Lab. On July 7, Reuters reported that the company had added Apple's (NASDAQ:AAPL) Ruoming Pang as its latest recruit.
Microsoft's (NASDAQ:MSFT) OpenAI partnership faced issues after OpenAI bought Windsurf, an AI coding firm. Disputes arose over Microsoft’s access to WindSurf’s IP and its stake in a restructured OpenAI.
Q2 was also marked by a shift to AI in hardware, robotics and edge applications.
Chipmakers Advanced Micro Devices (NASDAQ:AMD) and Google introduced specialized AI accelerators, a potential challenge to NVIDIA's nearly three year run as the dominant provider.
Notable developments in robotics included Google Cloud and Samsung Electronics' (KRX:005930) partnership, integrating Google Cloud's advanced generative AI technology into Samsung's new home AI companion robot, Ballie.
Data center operators like Amazon Web Services and Google Cloud also increased their infrastructure investments in the US as part of an effort to reduce reliance on foreign markets and secure long-term AI compute capacity.
Companies began testing or rolling out new AI agent capabilities, empowered by the Model Context Protocol from Anthropic. Major tech players, along with payment giants Visa (NYSE:V), Mastercard (NYSE:MA), Stripe, Block (NYSE:SQ) and PayPal (NASDAQ:PYPL), began adopting the Model Context Protocol to integrate seamless payment functionality directly into AI chatbots, moving beyond simple browse to full commerce.
Microsoft enhanced its GitHub Copilot offering with new coding agents capable of autonomous actions, while a handful of companies, including Dataiku, Databricks and Atlassian (NASDAQ:TEAM), introduced tools designed to build, deploy and manage autonomous systems for real-world enterprise applications.
On the quantum computing side, a paper published by researcher Craig Gidney for Google’s Quantum AI division suggests that a quantum computer could break a highly secure 2048 bit encryption, like the kind used for online banking, much faster than previously thought, requiring fewer than a million qubits.
Quantum computing firms later saw their shares spike following bullish comments from NVIDIA's Huang at his company’s Paris GTC conference. Before Huang’s comments, IBM (NYSE:IBM) announced its development of the world's first large-scale, error-free quantum computer, set to launch by 2029.
AI trends to watch in Q3
Q2 confirmed the AI cycle is evolving beyond text-based chatbots to hardware, embodiment and commercial uses.
While the Magnificent Seven still largely drove returns in Q2, there's an expectation that earnings growth will broaden out to other sectors. Picton Investments' 2025 mid-year update suggests that foundational model growth is encountering headwinds, with competition challenging the need for extensive capital expenditure.
Graph indicating that investor enthusiasm for AI stocks has recently "lost altitude."
Graph via Picton Investments.
However, the firm also suggests that this shift is redirecting the spotlight to real-world AI applications, leading to an expected acceleration of industrial adoption and the creation of new companies.
At this year’s Web Summit conference in May, panelists emphasized the critical role of strategic early stage investments when it comes to navigating the evolving AI landscape and identifying new opportunities.
“Our take is (that) AI is going to upend a lot of technology businesses. In the specific sense, I am of increasingly high conviction that authoring software is going to be more or less free, and that's going to shake up the topology of the software business market (in terms of) what makes sense and what's investable,” said Brett Gibson, managing partner at Initialized Capital, during a panel discussion on where AI investment is headed next.
He added that customizable software will ultimately allow for tailored solutions for virtually any need.
In H2, quantum computing could continue its shift from pure research into early stage commercialization.
Updates may come from firms like IonQ (NYSE:IONQ), which recently raised US$1 billion to expand quantum networking, as well as Quantinuum and PsiQuantum, which may reach technical milestones.
Meanwhile, D-Wave (NYSE:QBTS) is pivoting toward hybrid commercial models, which may offer continued proof of revenue from quantum optimization-as-a-service.
However, the outcome of ongoing trade negotiations between the US and the rest of the world could impact chip capacity and rare earths supply chains, constraining the growth of AI hardware stocks.
The Trump administration’s imposition of 25 percent tariffs on Japan and South Korea may pose a threat to semiconductor capacity and rare earths equipment imports critical for AI hardware.
“Both countries have been close partners on economic security matters and have a lot to offer the United States on priority matters like shipbuilding, semiconductors, critical minerals and energy cooperation,” Asia Society Policy Institute vice president Wendy Cutler said in response to the hikes.
Investor takeaway
The second quarter of the year confirmed an evolution in the AI landscape as the industry moves beyond theoretical discussions to real-world applications and critical infrastructure development.
While geopolitical tensions and concerns about job displacement may continue to present challenges, this pivot could set the stage for continued innovation and adaptation as the industry navigates both opportunities and complexities.
Don’t forget to follow us @INN_Technology for real-time news updates!
Securities Disclosure: I, Meagen Seatter, hold no direct investment interest in any company mentioned in this article.