[20250529] Trump's tariff action faces obstacles | Good news, good news, Musk finally quits government job | TEM shorted [Buduan Investment Notes]

1. A U.S. federal court (the U.S. Court of International Trade) on Wednesday blocked President Trump's 'Liberation Day' tariffs from taking effect, ruling that Trump's imposition of comprehensive tariffs under the International Emergency Economic Powers Act (IEEPA) on grounds such as trade deficits constituted an overreach of authority. After the federal court blocked the tariffs, the Trump administration immediately filed an appeal notice, indicating that this matter is far from settled.

I asked AI, and it suggested that the Trump administration might employ the following primary methods to circumvent the International Trade Court's ruling:

– Appeal to a higher court, seeking to overturn the ruling by citing historical precedents (e.g., the Yoshida case), which supports the president's authority to impose tariffs under similar laws (the Trading with the Enemy Act, the predecessor to IEEPA). If the appellate court or the Supreme Court accepts this precedent, Trump's tariffs could be deemed legal.

However, the Supreme Court has recently been cautious about the expansion of executive power, especially regarding the 'Major Questions Doctrine,' which requires explicit congressional authorization for significant economic or political decisions. If the court finds that the IEEPA does not explicitly authorize the president to impose broad tariffs, Trump may face defeat.

– Switch legal bases, such as using Section 232, Section 301, or Section 122 to continue imposing tariffs.

Section 232 of the Trade Expansion Act of 1962: Allows the president to impose tariffs on national security grounds. Trump used this provision during his first term to levy tariffs on steel and aluminum imports, and courts have generally been lenient toward such 'national security' justifications.

Section 301 of the Trade Act of 1974: Permits the U.S. Trade Representative (USTR), under presidential guidance, to impose tariffs on unfair trade practices. This provision was used for tariffs on China and, as it is executed by the USTR, may be subject to the Administrative Procedure Act (APA), but it remains a viable option.

Section 122 of the Trade Act of 1974: Allows the president to impose temporary tariffs (up to 15% for 150 days) during a balance-of-payments crisis. This was the basis used by the Nixon administration in 1971 and could be revived by the Trump administration.

– Seek congressional legislation to explicitly authorize the president to impose tariffs.

– Adjust the scope or implementation of tariffs, narrowing their impact or redefining the emergency.

– Leverage trade negotiations to alleviate tariff pressures through new agreements.

– Administrative delays or selective enforcement, maintaining some tariffs while awaiting appeal results.

– Mobilize public support, turning the legal dispute into a political issue. After all, we're just ordinary investors in U.S. stock market and dislike such uncertainty, which makes it hard to decide on stock holdings. However, even Buffett thinks 4.2 isn't panic-worthy enough for him to buy, showing that true investors aren't averse to volatility. Thus, we should not only accept it but also welcome potential adjustments, as long as we keep cash on hand for maneuvering.

2. Can Tesla continue to rise? After Musk previously reduced his time spent on DOGE and returned full-time to his main business, he has now announced he's quitting government work. While this was expected, it clearly shows the difference between being your own boss and working for the president. $Tesla(TSLA.US)

Musk struck a deal with Telegram, allowing Grok to integrate with Telegram and share profits. If users become Grok members via Telegram, profits will be split 50-50. However, Telegram emphasized that Grok is not authorized to use the platform's 1 billion users' content. Well.

3. We just discussed Tempus AI the other day—a company without profits. If you're really keen, take a small bet. Then it got shorted. Spruce Point's report cited reasons like minimal AI components and data embellishment. Now we're waiting for the company's response. Personally, I think the low AI ratio isn't a big deal; data utility matters more. Of course, its rally was driven by the belief that AI could positively impact databases. If you play with fire, you'll get burned. Let's wait and see. $Tempus AI(TEM.US)

– Management's controversial history
Spruce Point pointed out that founder Eric Lefkofsky and several board members were involved in companies with financial issues (this part is historical):

– Groupon (co-founded by Lefkofsky): Post-IPO, its stock plunged 95%, and it was required by the SEC to restate nearly 50% of revenue due to recognition issues, disclosing material weaknesses in internal controls.

– InnerWorkings (founded by Lefkofsky): The audit committee chair once served as its CEO. The company was later investigated for financial restatements and internal control problems.

– Echo Global Logistics: Post-IPO, it disclosed internal control deficiencies. Though eventually acquired at a premium, it long underperformed the industry index by 110%.

– AI capability and revenue quality concerns

– Minimal AI revenue: In 2024, only 2% of revenue (~$12.4M) came directly from AI applications. Despite rebranding as 'Tempus AI,' its main revenue still relies on traditional genetic testing (70%) and data licensing.

– Product demo contradictions: The report cited screenshots from promotional videos, highlighting inconsistencies in the AI system (e.g., patient age vs. sample timing errors), questioning its technical reliability.

– Financial manipulation concerns

– Circular transactions: SoftBank invested $200M pre-IPO, then co-founded a JV with Tempus (each contributing $95M). Tempus booked the JV's capital as revenue, accused of 'capital recycling' to inflate performance.

– Debt accounting boosting profits: In Google Cloud's $250M debt, Tempus recorded a $25.6M non-cash gain from debt principal modifications as an operating expense offset, directly lifting adjusted EBITDA. Excluding this, LTM EBITDA should be -$102.6M (vs. reported -$77M).

– Inflated client contracts: The disclosed $900M total contract value (TCV) includes $3M non-binding options and $22.4M 'near-impossible' milestone payments.

4. Why is there such high optimism about AI's potential to advance biotech? Woody's team Despot previously noted: Companies like Twist Bioscience CORP. $Twist Bioscience(TWST.US), Tempus AI Inc. $Tempus AI(TEM.US), 10x Genomics $10X Genomics(TXG.US), and Illumina Inc. $Illumina(ILMN.US) are developing increasingly powerful tools to sequence and analyze DNA faster and more cost-effectively.

Despot stated that AI-driven molecular diagnostics are transforming early disease detection and enabling precise classification of individual patients' biological traits (this is somewhat like the 'observation' in traditional Chinese medicine—face-reading, haha).

He said, 'AI leverages data from multi-omics tools and diagnostics to design more effective drugs, delivering higher success rates faster and at lower cost.' (No argument there.)

– So next, we need to spot biotech and data companies that truly benefit from and actively use AI. As with many AI firms, profitability may take time.

5. OpenAI is restructuring, reportedly paving the way for a future IPO.