“Taiwan will be the most sensitive issue raised at the meeting,” ASPI analyst Nathan Attrill warned — and that sensitivity threads through a summit where trade, rare-earths controls, nuclear weapons, Taiwan, Iran, artificial-intelligence risks and export controls on semiconductors are all on the table. Presidents Donald Trump and Xi Jinping meet in Beijing on May 14–15 for the first time since 2017 with a short, sharp agenda and a long list of potential second‑order effects for allies, markets and technologies.
Taiwan: ambiguity is the strategic risk
Nathan Attrill says Beijing will press hard on arms sales and official contact with Taiwan, while Washington is “unlikely to make a clear public concession.” The more likely immediate outcome, he predicts, is tactical: a reduction in high‑profile Chinese military activity around the summit and continued US arms support afterward. That pattern — a pause for optics followed by a return to business as usual — carries its own danger. Attrill writes that “the risk is that Beijing reads any delay, softer language or transactional framing as a sign that US support for Taiwan is negotiable,” and that this could “encourage more pressure on Taiwan after the summit.”
The US agenda: trade, Iran and the domestic sell
Bethany Allen outlines what the White House is seeking: dealable, headline‑friendly outcomes that can be marketed to a domestic audience. Expect announcements such as purchases of agricultural products and “the establishment of a forum to discuss trade and investment.” The war with Iran will be prominent in public remarks — the United States will seek China’s help “in opening the Strait of Hormuz and bringing the conflict to a close” — but Allen counsels restraint about expectations: “Don’t expect breakthroughs on Iran,” because allies have largely refused the same calls and China “has even less incentive to expend its own diplomatic and military capital” to achieve US objectives.
AI and chips: competition with narrow cooperation
David Wroe frames artificial intelligence as “the ultimate strategic prize,” with intense competition persisting even as limited cooperation may be possible. In the wake of Anthropic’s announcement of its Mythos model, Wroe suggests the summit is a chance to agree on “minimum safety standards for frontier models” or non‑binding practices such as threat‑actor information sharing and an “AI‑crisis hotline.” At the same time, he reports Xi’s ask: that Trump wind back US export controls on advanced chips used for training and running AI models. Wroe’s prescription is blunt — “Trump should refuse” — because better chips underpin US advantage in AI, according to his analysis.
Hybrid coercion, cyber threats and human rights
Fitriani warns the summit will likely prioritise transactional issues — trade, Iran, Taiwan — over deeper conversations about hybrid coercive tactics that threaten allies and partners. She lists cyber intrusions into critical infrastructure, maritime grey‑zone activities, information operations, strategic dependence on rare earths, and the weaponisation of dual‑use technologies as growing global threats that are unlikely to be constrained at the meeting. Tilla Hoja adds that human rights will probably be downplayed: any diminution of pressure on issues such as the treatment of Uyghurs in Xinjiang “will be seen as a win for Xi.” Yet Trump has signalled he will raise the jailing of Jimmy Lai, the Hong Kong media tycoon sentenced to 20 years under Hong Kong’s 2020 national security law — a narrow nod toward democratic values amid a summit focused on transactional diplomacy.
How Taiwan, Japan, AI researchers, and Hong Kong advocates are likely to react
- Taiwan: Watch official language. Taiwanese officials will scrutinise whether post‑summit statements treat Taiwan as “a core security partner” or “a problem to be managed,” and will worry that ambiguity could invite renewed pressure.
- Japan: Tokyo will be “wary of discussions about Taiwan,” given the island’s and Okinawa’s role in Japan’s maritime security; meetings with US Treasury Secretary Scott Bessent in Tokyo this week sought to impress Japan’s China policy on US interlocutors to avert an unfavourable outcome.
- AI researchers and security teams: They will look for any agreement on minimum safety standards, threat‑sharing mechanisms, or an AI‑crisis hotline — tangible, technical measures that could reduce common risks even as strategic rivalry continues.
- Hong Kong and human‑rights advocates: They will test Washington’s resolve based on whether Trump presses the issue of Jimmy Lai and how visible human‑rights language is in public statements after the visit.
The summit is therefore both narrowly choreographed and broadly consequential. Expect headline‑driven deliverables — agricultural purchases, a trade forum, public pledges of cooperation on some security questions — and persistent, underlying fault lines: Taiwan’s status, export controls on semiconductors, rare‑earths leverage, hybrid coercion, and tightly contested AI standards. The immediate markers to watch, in the order analysts themselves put weight on them, are the language used about Taiwan, any announced trade measures and forums, whether AI‑related minimum standards or hotlines are established, and whether human‑rights concerns such as Jimmy Lai’s case receive substantive attention. Those outcomes will tell whether the meeting is a managed pause in competition, a transactional bargain with regional consequences, or something more durable.
https://www.aspistrategist.org.au/mr-trump-goes-to-beijing-views-from-aspi-analysts/
