🚀 6 million registered for Trump Accounts

Market Overview
Read time 1.4 minutes

Year To Date Performances:

1. More than 6 million American children have been registered for Trump Accounts ahead of the program’s official July 4 launch, according to recent U.S. Department of the Treasury figures. However, policy experts warn that millions of eligible youth are still missing out, with lower-income households significantly underrepresented. While any U.S. citizen under age 18 with a Social Security number is eligible for the tax-deferred investing account, only 1.4 million enrollees have been cleared for the Treasury’s one-time $1,000 pilot contribution for children born between 2025 and 2028—a figure representing just 39% of eligible infants nationwide. The opt-in framework, which requires parents to file IRS Form 4547 or register online at TrumpAccounts.gov, has created an accessibility gap: families earning over $200,000 are statistically overrepresented among early sign-ups, while lower-income households are less likely to file federal tax returns regularly. To broaden participation, the initiative features a "50-state challenge" championed by Treasury Secretary Scott Bessent to secure local philanthropic matches, highlighted by a $6.25 billion pledge from Michael and Susan Dell to seed an extra $250 for eligible children born between 2016 and 2024. Despite these incentives, think tanks like the Cato Institute caution that the accounts' complex distribution rules—which blend pre-tax and after-tax dollars subject to ordinary income taxes and a 10% early withdrawal penalty before age 59½—may continue to deter the financially strained families who stand to benefit the most from long-term compounding.

2. Despite its historical dominance, Google's core search business is showing unprecedented vulnerabilities in the artificial intelligence era, evidenced by a 5% drop in Alphabet's stock following a high-profile talent defection and an intensifying user backlash against automated search results. While the tech giant still controls 90% of the search market, recent metrics indicate a 1% decline in Google's search engine traffic, contrasted by an increase for OpenAI's ChatGPT—which recently hit 1 billion monthly active users—and a milestone 1 billion users for Microsoft's Bing. This shifting landscape has triggered a dual challenge for Google: it risks cannibalizing its lucrative digital advertising model, which accounts for three-quarters of its total revenue, while simultaneously driving users toward privacy-focused and "AI-free" alternatives. DuckDuckGo has capitalized on this fatigue by introducing a "no-AI" search engine variant, driving their extension installation rates up by as much as 75% following Google's recent I/O conference announcement, where an "AI Mode" button was integrated directly into the iconic 25-year-old search box. Compounding these retention anxieties, digital publishers are bracing for a future independent of search traffic, as data panels reveal that roughly 68% of Google searches now resolve as "zero-click" experiences due to AI Overviews automatically scraping content into on-page summaries. To defend its ecosystem, Google has aggressively accelerated capital infrastructure spending toward a projected $200 billion; however, the enterprise faces severe headwinds in the ongoing war for AI talent, underscored by the high-profile losses of Gemini co-lead Noam Shazeer to OpenAI and DeepMind fellow John Jumper to Anthropic in a single week.

3. Eight-month-old AI memory startup Engram has secured $98 million in a funding round led by General Catalyst, Kleiner Perkins, and Sequoia Capital, alongside prominent angel investor Andrej Karpathy, to address the escalating token overhead plaguing commercial developer operations. Founded by computational neuroscientist and former Stanford AI researcher Dan Biderman, the 13-person enterprise deploys "learned memory" architectures that map organization-specific workflows, allowing models to retain historical context and accurately anticipate user queries while using up to 100 times fewer tokens than frontier foundational models. The startup's context-optimization technology arrives as corporate developers increasingly crack down on runaway inference costs, enabling Engram to quickly assemble an elite roster of enterprise clients that includes Microsoft, Notion, and legal-tech unicorn Harvey. By intentionally prioritizing hyper-specialized domain intuition over broad general-knowledge capabilities, Engram aims to utilize its new capital to scale its engineering talent and computing infrastructure, turning enterprise context into an efficient layer of intuition that prevents businesses from overpaying for redundant, data-heavy prompt processing.

   Headlines

   1. Oracle has cut staff by 21,000 over the last year, replacing jobs with AI.

   2. SpaceX has lost over $400B from its peak value.