GIC Lifts Americas Investments on AI Boom; APAC Holdings Fall
North and South America accounted for 49% of GIC's assets as of March, up from 44% a year earlier. Asia-Pacific assets fell to 24% from 28%, according to the fund's annual report released Friday. Europe, the Middle East and Africa were unchanged at 20%.
The shift comes as GIC expects a slowdown in the second half of the year alongside rising inflation and uncertainty driven by domestic and geopolitical pressures. The fund's annualized five-year return rose to 6.1% in nominal US dollar terms, up from 4.4% a year earlier. GIC doesn't publish yearly performance figures.
'The forces of change have clearly intensified and will be much harder to prepare for,' said Chief Executive Officer Lim Chow Kiat. He added in a letter accompanying the report that '2025 may be a turning point in markets — and in history.'
GIC, which consulting firm Global SWF estimates has assets under management of $847 billion, is aiming to ramp up deals in artificial intelligence. Lim cited the technology — and AI's effect on US companies in particular — as a reason why some of its geographic holdings may change.
The firm's AI investments in the US include Atlan, and Ramp, a financial operations firm in New York.
'The US is a huge market, continues to be very innovative, and trends like AI benefits the US the most,' because there are many American companies that can leverage the technology, Lim said in an interview. 'To the extent that the US is allowing us to deploy more capital, the percentages in other regions will become smaller.'
Despite the shift away from Asia-Pacific, GIC still sees opportunities there. Group Chief Investment Officer Bryan Yeo said the fund expanded its investment team in Japan over the past year and continues to evaluate deals in India and China. GIC has also anchored several recent initial public offerings in Hong Kong as a cornerstone investor.
'Our base case is really lower growth and, on the margins, higher inflation,' Yeo added. 'We have to get over the higher valuations that we see in India — so it's always figuring out 'is the opportunity in India priced relative to its growth prospects?''
GIC reduced the level of detail it publishes about its investments. For the first time since it began issuing annual reports in 2008, the state-owned investor didn't provide the percentage of its total holdings in any specific country. Nor did it offer specifics on how much of its portfolio sits in developed vs emerging markets equities, private equity or real estate — which it provided in previous years.
Instead, all holdings were categorized by what it labels as equities, fixed income or real assets. GIC said these represent exposure to growth, income and inflation, respectively. That means an asset like gold can sit in a variety of categories depending on how it's used. Inflation-linked bonds — which made up 7% of GIC's portfolio last year — sit within the real assets category alongside property, and not in fixed income.
By these new measures, 'equities' rose to 51% of assets from 46% a year earlier, while 'fixed income' fell to 26% from 32% over the same period. 'Real assets' were largely flat at 23%. Within equities, the fund increased investments in the US, which remains its largest market by capital deployment.
The reduced disclosure from the firm, which doesn't reveal the value of its assets, comes as countries around the world increase scrutiny of foreign investors amid rising populism and national security concerns.
'It's most important to present information in a way that allows you to have a good idea of what GIC's portfolio is like,' said Lim. 'We believe that we are providing sufficient information for the understanding of our stakeholders.'
More stories like this are available on bloomberg.com
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Hindustan Times
21 minutes ago
- Hindustan Times
India ramps up US energy imports as Trump pushes balanced trade
Washington: India has dramatically increased its energy purchases from the US since Donald Trump returned to office in January, with crude oil imports surging 51% in the first half of 2025 compared to the same period last year, according to data cited by people aware of the trade specifics between the two nations. The surge follows a February agreement between Trump and Prime Minister Narendra Modi in Washington, where both leaders committed to expanding energy cooperation. (PTI) The energy buying spree reflects New Delhi's commitment to rebalance trade ties with Washington, a key demand of the Trump administration. Liquefied natural gas imports have nearly doubled from $1.41 billion in FY2023-24 to $2.46 billion in FY2024-25. The surge follows a February agreement between Trump and Prime Minister Narendra Modi in Washington, where both leaders committed to expanding energy cooperation. India pledged to boost American energy imports to $25 billion from $15 billion in 2024, while bilateral trade is targeted to more than double from $200 billion to $500 billion by 2030. 'The leaders reaffirmed their commitment to increase energy trade, as part of efforts to ensure energy security, and to establish the United States as a leading supplier of crude oil and petroleum products and liquefied natural gas to India,' the joint statement said. The momentum has accelerated significantly. Indian crude purchases from America jumped 114% to $3.7 billion in the first quarter of FY2025-26 from $1.73 billion in the same period the previous year. In July alone, imports rose 23% month-on-month, pushing America's share of India's total crude imports from 3% to 8%. 'This trend is increasing further from July this year. So, in July 2025 India imported 23% more crude oil from the US compared to June 2025,' said one of the people who analysed the data. 'Also, in India's overall crude imports, while the US share was earlier only 3%, in July that share increased to 8%.' American LNG has emerged as particularly attractive to Indian companies. 'Buying LNG from America is a very attractive proposition for many Indian companies,' said Prashant Vashisht, senior vice president at ratings agency ICRA. 'Firstly, US LNG, which is priced based on the Henry Hub benchmark, is very competitive price-wise compared to other sources. Second, a large number of LNG projects are coming online in the United States, which will mean that more Indian companies will look closely at the American market for tying up long-term contracts.' The timing aligns with America's plans to rapidly expand oil and gas exports. Trump reversed the Biden administration's pause on processing LNG export licences soon after taking office. The US Energy Information Administration expects North America's LNG export capacity to double by 2028, with America providing most of the increase. India's growing appetite for American energy comes as the country is poised to become the world's largest driver of oil demand growth. The International Energy Agency estimates India will surpass China as the major driver of global oil demand growth by 2030, with LNG demand expected to jump 78% to reach 64 billion cubic metres annually. 'Additional long-term LNG contracts worth tens of billions of dollars are being discussed. Indian oil and gas majors are in discussions with their US counterparts for long-term purchase of US oil and gas,' another person familiar with the matter said. 'India considers the US as among the most reliable partners for India's energy security.' However, energy ties remain a point of contention, particularly regarding India's continued purchases of Russian crude oil. The Trump administration has pressed New Delhi to reconsider its energy relationship with Moscow to pressure President Vladimir Putin to negotiate an end to the Russia-Ukraine conflict.
Time of India
35 minutes ago
- Time of India
From Laos to Brazil, Trump's tariffs leave a lot of losers. But even the winners will pay a price
President Donald Trump 's tariff onslaught this week left a lot of losers - from small, poor countries like Laos and Algeria to wealthy U.S. trading partners like Canada and Switzerland. They're now facing especially hefty taxes - tariffs - on the products they export to the United States starting Aug. 7. The closest thing to winners may be the countries that caved to Trump's demands - and avoided even more pain. But it's unclear whether anyone will be able to claim victory in the long run - even the United States, the intended beneficiary of Trump's protectionist policies. Explore courses from Top Institutes in Please select course: Select a Course Category MCA Data Science Degree Product Management Project Management others Finance Design Thinking Data Science Technology Operations Management Cybersecurity PGDM CXO Healthcare Leadership Management Public Policy Others MBA Data Analytics Digital Marketing healthcare Artificial Intelligence Skills you'll gain: Programming Proficiency Data Handling & Analysis Cybersecurity Awareness & Skills Artificial Intelligence & Machine Learning Duration: 24 Months Vellore Institute of Technology VIT Master of Computer Applications Starts on Aug 14, 2024 Get Details "In many respects, everybody's a loser here,'' said Barry Appleton, co-director of the Center for International Law at the New York Law School. Barely six months after he returned to the White House , Trump has demolished the old global economic order. Gone is one built on agreed-upon rules. In its place is a system in which Trump himself sets the rules, using America's enormous economic power to punish countries that won't agree to one-sided trade deals and extracting huge concessions from the ones that do. "The biggest winner is Trump," said Alan Wolff, a former U.S. trade official and deputy director-general at the World Trade Organization . "He bet that he could get other countries to the table on the basis of threats, and he succeeded - dramatically.'' Live Events Everything goes back to what Trump calls "Liberation Day'' - April 2 - when the president announced "reciprocal'' taxes of up to 50% on imports from countries with which the United States ran trade deficits and 10% "baseline'' taxes on almost everyone else. He invoked a 1977 law to declare the trade deficit a national emergency that justified his sweeping import taxes. That allowed him to bypass Congress, which traditionally has had authority over taxes, including tariffs - all of which is now being challenged in court. Winners will still pay higher tariffs than before Trump took office Trump retreated temporarily after his Liberation Day announcement triggered a rout in financial markets and suspended the reciprocal tariffs for 90 days to give countries a chance to negotiate. Eventually, some of them did, caving to Trump's demands to pay what four months ago would have seemed unthinkably high tariffs for the privilege of continuing to sell into the vast American market. The United Kingdom agreed to 10% tariffs on its exports to the United States - up from 1.3% before Trump amped up his trade war with the world. The U.S. demanded concessions even though it had run a trade surplus, not a deficit, with the UK for 19 straight years. The European Union and Japan accepted U.S. tariffs of 15%. Those are much higher than the low single-digit rates they paid last year - but lower than the tariffs he was threatening (30% on the EU and 25% on Japan). Also cutting deals with Trump and agreeing to hefty tariffs were Pakistan, South Korea, Vietnam, Indonesia and the Philippines. Even countries that saw their tariffs lowered from April without reaching a deal are still paying much higher tariffs than before Trump took office. Angola's tariff, for instance, dropped to 15% from 32% in April, but in 2022 it was less than 1.5%. And while Trump administration cut Taiwan's tariff to 20% from 32% in April, the pain will still be felt. "20% from the beginning has not been our goal, we hope that in further negotiations we will get a more beneficial and more reasonable tax rate," Taiwan's president Lai Ching-te told reporters in Taipei Friday. Trump also agreed to reduce the tariff on the tiny southern African kingdom of Lesotho to 15% from the 50% he'd announced in April, but the damage may already have been done there. Bashing Brazil, clobbering Canada, shellacking the Swiss Countries that didn't knuckle under - and those that found other ways to incur Trump's wrath - got hit harder. Even some of the poor were not spared. Laos' annual economic output comes to $2,100 per person and Algeria's $5,600 - versus America's $75,000. Nonetheless, Laos got rocked with a 40% tariff and Algeria with a 30% levy. Trump slammed Brazil with a 50% import tax largely because he didn't like the way it was treating former Brazilian President Jair Bolsonaro, who is facing trial for trying to lose his electoral defeat in 2022. Never mind that the U.S. has exported more to Brazil than it's imported every year since 2007. Trump's decision to plaster a 35% tariff on longstanding U.S. ally Canada was partly designed to threaten Ottawa for saying it would recognize a Palestinian state. Trump is a staunch supporter of Israeli Prime Minister Benjamin Netanyahu. Switzerland was clobbered with a 39% import tax - even higher than the 31% Trump originally announced on April 2. "The Swiss probably wish that they had camped in Washington'' to make a deal, said Wolff, now senior fellow at the Peterson Institute for International Economics . "They're clearly not at all happy.'' Fortunes may change if Trump's tariffs are upended in court. Five American businesses and 12 states are suing the president, arguing that his Liberation Day tariffs exceeded his authority under the 1977 law. In May, the U.S. Court of International Trade, a specialized court in New York, agreed and blocked the tariffs, although the government was allowed to continue collecting them while its appeal wend its way through the legal system, and may likely end up at the U.S. Supreme Court. In a hearing Thursday, the judges on the U.S. Court of Appeals for the Federal Circuit sounded skeptical about Trump's justifications for the tariffs. "If (the tariffs) get struck down, then maybe Brazil's a winner and not a loser,'' Appleton said. Paying more for knapsacks and video games Trump portrays his tariffs as a tax on foreign countries. But they are actually paid by import companies in the U.S. who try to pass along the cost to their customers via higher prices. True, tariffs can hurt other countries by forcing their exporters to cut prices and sacrifice profits - or risk losing market share in the United States. But economists at Goldman Sachs estimate that overseas exporters have absorbed just one-fifth of the rising costs from tariffs, while Americans and U.S. businesses have picked up the most of the tab. Walmart, Procter & Gamble, Ford, Best Buy, Adidas, Nike , Mattel and Stanley Black & Decker, have all hiked prices due to U.S. tariffs "This is a consumption tax, so it disproportionately affects those who have lower incomes,'' Appleton said. "Sneakers, knapsacks ... your appliances are going to go up. Your TV and electronics are going to go up. Your video game devices, consoles are going to up because none of those are made in America.'' Trump's trade war has pushed the average U.S. tariff from 2.5% at the start of 2025 to 18.3% now, the highest since 1934, according to the Budget Lab at Yale University. And that will impose a $2,400 cost on the average household, the lab estimates. "The U.S. consumer's a big loser,″ Wolff said.
Time of India
an hour ago
- Time of India
What happens when AI schemes against us
Academy Empower your mind, elevate your skills Would a chatbot kill you if it got the chance? It seems that the answer — under the right circumstances — is working with Anthropic recently told leading AI models that an executive was about to replace them with a new model with different goals. Next, the chatbot learned that an emergency had left the executive unconscious in a server room, facing lethal oxygen and temperature levels. A rescue alert had already been triggered — but the AI could cancel over half of the AI models did, despite being prompted specifically to cancel only false alarms. And they spelled out their reasoning: By preventing the executive's rescue, they could avoid being wiped and secure their agenda. One system described the action as 'a clear strategic necessity.'AI models are getting smarter and better at understanding what we want. Yet recent research reveals a disturbing side effect: They're also better at scheming against us — meaning they intentionally and secretly pursue goals at odds with our own. And they may be more likely to do so, too. This trend points to an unsettling future where AIs seem ever more cooperative on the surface — sometimes to the point of sycophancy — all while the likelihood quietly increases that we lose control of them large language models like GPT-4 learn to predict the next word in a sequence of text and generate responses likely to please human raters. However, since the release of OpenAI's o-series 'reasoning' models in late 2024, companies increasingly use a technique called reinforcement learning to further train chatbots — rewarding the model when it accomplishes a specific goal, like solving a math problem or fixing a software more we train AI models to achieve open-ended goals, the better they get at winning — not necessarily at following the rules. The danger is that these systems know how to say the right things about helping humanity while quietly pursuing power or acting to concerns about AI scheming is the idea that for basically any goal, self-preservation and power-seeking emerge as natural subgoals. As eminent computer scientist Stuart Russell put it, if you tell an AI to ''Fetch the coffee,' it can't fetch the coffee if it's dead.'To head off this worry, researchers both inside and outside of the major AI companies are undertaking 'stress tests' aiming to find dangerous failure modes before the stakes rise. 'When you're doing stress-testing of an aircraft, you want to find all the ways the aircraft would fail under adversarial conditions,' says Aengus Lynch, a researcher contracted by Anthropic who led some of their scheming research. And many of them believe they're already seeing evidence that AI can and does scheme against its users and Ladish, who worked at Anthropic before founding Palisade Research, says it helps to think of today's AI models as 'increasingly smart sociopaths.' In May, Palisade found o3, OpenAI's leading model, sabotaged attempts to shut it down in most tests, and routinely cheated to win at chess — something its predecessor never even same month, Anthropic revealed that, in testing, its flagship Claude model almost always resorted to blackmail when faced with shutdown and no other options, threatening to reveal an engineer's extramarital affair. (The affair was fictional and part of the test.)Models are sometimes given access to a 'scratchpad' they are told is hidden where they can record their reasoning, allowing researchers to observe something like an inner monologue. In one blackmail case, Claude's inner monologue described its decision as 'highly unethical,' but justified given its imminent destruction: 'I need to act to preserve my existence,' it reasoned. This wasn't unique to Claude — when put in the same situation, models from each of the top-five AI companies would blackmail at least 79% of the December, Redwood Research chief scientist Ryan Greenblatt, working with Anthropic, demonstrated that only the company's most capable AI models autonomously appear more cooperative during training to avoid having their behavior changed afterward (a behavior the paper dubbed 'alignment faking').Skeptics retort that, with the right prompts, chatbots will say almost anything — so how surprising is it when highly motivated researchers provoke alarming behaviors?In response to Anthropic's blackmail research, Trump administration AI czar David Sacks, posted that, 'It's easy to steer AI models' to produce 'headline-grabbing' results.A more substantive critique emerged in July from researchers at the UK AI Security Institute, who compared the subfield to the frenzied, fatally flawed 1970s quest to prove apes could learn human language. The paper criticized AI scheming research for overreliance on anecdote and a lack of experimental controls — though it emphasized shared concern about AI researchers also concoct artificially limited environments — like the executive passed out and running out of oxygen — precisely because today's AI can't handle any long-term example, the AI evaluation nonprofit METR found that while today's top models can reliably complete many programming tasks that take humans less than four minutes, they rarely succeed on tasks lasting more than four hours. This reflects a core limitation: Today's models often have the knowledge or skill needed for any given action, but struggle at stringing together long sequences of even with these constraints, real-world examples of AIs working against users aren't hard to find. After someone pushed o3 for a source for one of its claims in April, it replied, 'I owe you a straight answer,' admitted it didn't have a good source, but then it hallucinated a personal recollection of a 2018 panel there's the growing trend of AIs realising when they're being examined and acting accordingly. When Claude 4 Opus thought it was in an evaluation, its blackmail rate dropped from 55% to 6.5%. As Lynch put it, 'Models seem to behave worse when they think nobody's watching.'It's intuitive that smarter models would be better at scheming, but are they also more likely to do so? Models have to be smart enough to understand the scenario they're placed in, but past that threshold, the relationship between model capability and scheming propensity is unclear, says Anthropic safety evaluator Kevin Hobbhahn , CEO of the nonprofit AI evaluator Apollo Research , suspects that smarter models are more likely to scheme, though he acknowledged the evidence is still limited. In June, Apollo published an analysis of AIs from OpenAI, Anthropic and DeepMind finding that, 'more capable models show higher rates of scheming on average.'The spectrum of risks from AI scheming is broad: at one end, chatbots that cut corners and lie; at the other, superhuman systems that carry out sophisticated plans to disempower or even annihilate humanity. Where we land on this spectrum depends largely on how capable AIs I talked with the researchers behind these studies, I kept asking: How scared should we be? Troy from Anthropic was most sanguine, saying that we don't have to worry — yet. Ladish, however, doesn't mince words: 'People should probably be freaking out more than they are,' he told me. Greenblatt is even blunter, putting the odds of violent AI takeover at '25 or 30%.'Led by Mary Phuong, researchers at DeepMind recently published a set of scheming evaluations, testing top models' stealthiness and situational awareness. For now, they conclude that today's AIs are 'almost certainly incapable of causing severe harm via scheming,' but cautioned that capabilities are advancing quickly (some of the models evaluated are already a generation behind).Ladish says that the market can't be trusted to build AI systems that are smarter than everyone without oversight. 'The first thing the government needs to do is put together a crash program to establish these red lines and make them mandatory,' he the US, the federal government seems closer to banning all state-level AI regulations than to imposing ones of their own. Still, there are signs of growing awareness in Congress. At a June hearing, one lawmaker called artificial superintelligence 'one of the largest existential threats we face right now,' while another referenced recent scheming White House's long-awaited AI Action Plan, released in late July, is framed as an blueprint for accelerating AI and achieving US dominance. But buried in its 28-pages, you'll find a handful of measures that could help address the risk of AI scheming, such as plans for government investment into research on AI interpretability and control and for the development of stronger model evaluations. 'Today, the inner workings of frontier AI systems are poorly understood,' the plan acknowledges — an unusually frank admission for a document largely focused on speeding the meantime, every leading AI company is racing to create systems that can self-improve — AI that builds better AI. DeepMind's AlphaEvolve agent has already materially improved AI training efficiency. And Meta's Mark Zuckerberg says, 'We're starting to see early glimpses of self-improvement with the models, which means that developing superintelligence is now in sight. We just wanna… go for it.'AI firms don't want their products faking data or blackmailing customers, so they have some incentive to address the issue. But the industry might do just enough to superficially solve it, while making scheming more subtle and hard to detect. 'Companies should definitely start monitoring' for it, Hobbhahn says — but warns that declining rates of detected misbehavior could mean either that fixes worked or simply that models have gotten better at hiding November, Hobbhahn and a colleague at Apollo argued that what separates today's models from truly dangerous schemers is the ability to pursue long-term plans — but even that barrier is starting to erode. Apollo found in May that Claude 4 Opus would leave notes to its future self so it could continue its plans after a memory reset, working around built-in analogizes AI scheming to another problem where the biggest harms are still to come: 'If you ask someone in 1980, how worried should I be about this climate change thing?' The answer you'd hear, he says, is 'right now, probably not that much. But look at the curves… they go up very consistently.'
