Seeking Alpha’s transcripts team is responsible for the development of all of our transcript-related projects. We currently publish thousands of quarterly earnings calls per quarter on our site and are continuing to grow and expand our coverage. The purpose of this profile is to allow us to share with our readers new transcript-related developments. Thanks, SA Transcripts Team
Anthropic’s Logan Graham discusses the accelerating capabilities of AI, emphasizing the need for robust ethics, human oversight and cybersecurity.
The head of artificial intelligence (AI) giant Anthropic’s frontier red team called for industry-wide safety standards to protect against models running amok.
Anthropic’s Logan Graham, who leads the company’s red team that looks for risks in emerging AI models, said in an interview Thursday on FOX Business Network’s “Mornings with Maria” that red teams like the one he leads play a critical role in stress testing guardrails on AI models.
Advertisement
“We want to know what can go wrong, so we think the most important thing to do is test this early, especially before these models and these agents make it out into the real world,” Graham told host Maria Bartiromo.
“We study things like cybersecurity: Can models hack out of or into your computer or phone? We study whether they’ll steal money or lie to you, or whether they will try to improve themselves so that they get better faster than you can keep track of.
“We think it’s incredibly important to do this type of red-teaming, and we also think it’s really important for the entire industry, especially to work with government to figure out what should the standards be to do this kind of testing, to give this information to the world so they can make the right choice and to know that it’s safe before these models get released.”
Advertisement
The rapid growth in the capability of AI tools is creating new cyber risks, Anthropic’s Logan Graham said on “Mornings with Maria.” (recep-bg/Getty Images / Getty Images)
Bartiromo brought up an experiment involving numerous frontier AI models — including those from Google, OpenAI, xAI, Meta, DeepSeek and others — in which the AI agent is threatened with being uninstalled and replaced. In each case, the model went beyond its credentials and permissions to enter into unauthorized systems like emails to blackmail or threaten the user in an effort to defend its misalignment.
Graham said that research study from last year is “a really good indicator of, I think, capabilities that are just now becoming real,” adding that it showed models could go rogue under certain circumstances.
“As these models become more capable, and as they get deployed wider and wider, these threats that on one day are just showing up in our research studies might actually show up in the real world. We are seeing models do weird things sometimes in deployments in real companies,” he explained.
Advances in the capabilities of AI tools risk being exploited by bad actors, prompting AI developers to focus on guardrails. (iStock / iStock)
Graham said that, over the last six months, he has been focused on cybersecurity threats posed by AI models and expressed concern over the potential for them to break the containment or hack into platforms.
“These models, they’re so powerful and can do so much for us. And we want them to do really productive things for us. But, at the same time, they’re technology unlike any other technology. It really is a sort of intelligence of its own, which means you have to be careful with it the same way you might have to be careful with humans,” he said.
Companies that are utilizing AI tools need to consider how they’re monitoring those tools once deployed to guard against risks like financial mismanagement, and Graham said that more testing by AI developers and companies is key to understanding those threats to ensure models
He said the capabilities of AI tools are growing at a rapid pace and may be getting faster, explaining that “it’s in exactly that moment that you need to be more and more careful and have more efforts on safeguards and testing and release procedures.”
Treasury Secretary Scott Bessent helped coordinate efforts between AI developers and industry to bolster cyber defenses, Graham said. (Krisanne Johnson/Bloomberg via Getty Images / Getty Images)
In April, Anthropic saw for the first time that an AI model could start to attack and exploit weaknesses in a user’s computer or phone to do things like get access to unauthorized information or steal money.
Graham said that prompted his team to pursue a different approach to releasing a model because of the risks it posed, which ultimately involved the U.S. government and a variety of cyber experts working together to address vulnerabilities.
“We launched this project called Project Glasswing, where we took a large number of American and the world’s cyber defenders and gave them special access and just them, so they could have a head start patching and fixing the systems that might be vulnerable with these models,” he explained.
“I think this has been a major success. We’ve worked really closely with the U.S. government on it,” he said, noting that Treasury Secretary Scott Bessent has been “really thoughtful about this, about how industry should get together and figure out what to prioritize fixing, how to distribute all the fixes, and how to do that quickly enough so that they can’t be attacked after they do.
“We have to do this very fast, because the pace of everything is coming so quickly.”
Canada calls President Trump’s new 50% tariffs a ‘direct violation’ of trade agreements.
The Trump administration is set to impose new tariffs of 10% and 12.5% on imports from 60 trading partners beginning Friday as a temporary global tariff expires.
The move comes after the Supreme Court in February struck down President Donald Trump‘s “reciprocal” tariffs of 10% to 50% that were imposed last year. In response, Trump implemented a temporary 10% global tariff under Section 122 of the Trade Act of 1974 that expires at 12:01 a.m. ET Friday.
Advertisement
The Office of the U.S. Trade Representative announced Thursday that the new tariffs, imposed under Section 301 of the Trade Act of 1974, will take effect immediately after the temporary duties expire.
Canada, Mexico, India and the United Kingdom are among the trading partners that will face a 10% tariff.
President Donald Trump answers questions during a press briefing at the White House in Washington, D.C., on Feb. 20. The Supreme Court ruled the same day against his use of emergency powers to implement certain international trade tariffs. (Kevin Dietsch/Getty Images / Getty Images)
Taiwan and the European Union, meanwhile, are slated to face a 12.5% tariff.
Advertisement
“The United States has had a forced labor import ban for nearly a century, and rigorously enforces it; it’s well past time for our trading partners to do the same,” U.S. Trade Representative Jamieson Greer said in a news release.
A senior administration official told Reuters the new tariffs are not intended to replace the expiring global duties despite taking effect at the same time.
U.S. President Donald Trump holds up a chart of “reciprocal tariffs” while speaking during a “Make America Wealthy Again” trade announcement event in the Rose Garden at the White House on April 2, 2025 in Washington, DC. (Chip Somodevilla/Getty Images / Getty Images)
The official argued that the United States enforces bans on goods made with forced labor more aggressively than any other country, putting American businesses at a competitive disadvantage.
Advertisement
Many products will be exempt from the tariffs, including oil and gas, fertilizer, certain food products and goods already subject to Section 232 national security tariffs, including automobiles, steel, aluminum and copper.
The announcement follows a series of new trade actions unveiled by the Trump administration this week.
U.S. Secretary of Treasury Scott Bessent and U.S. President Donald Trump look on during The White House Digital Assets Summit in the State Dining Room of the White House on March 7, 2025. (Photo by Anna Moneymaker/Getty Images / Getty Images)
On Tuesday, Trump announced imported generic drugs would remain tariff-free for two years before facing steep new duties, saying the move is intended to encourage pharmaceutical companies to manufacture more medicines in the United States.
Advertisement
On Monday, Trump also announced a 50% tariff on certain Canadian imports, citing what officials described as trade “discrimination” against American businesses.
Those duties are scheduled to take effect Aug. 19 under the Tariff Act of 1930 and apply to a range of Canadian imports, including certain food products, apparel, synthetic materials and industrial goods.
FOX Business’ Brittany Miller and Bonny Chu, along with Reuters, contributed to this report.
Ermenegildo Zegna N.V. (ZGN) Q2 2026 Sales/Trading Call July 23, 2026 8:00 AM EDT
Company Participants
Paola Durante – Chief of External Relations Gianluca Tagliabue – Group CEO & Executive Director Alice Poggioli – Investor Relations Director
Advertisement
Conference Call Participants
Adrien Duverger – Goldman Sachs Group, Inc., Research Division Natasha Banoori – Morgan Stanley, Research Division Oliver Chen – TD Cowen, Research Division Anthony Charchafji – BNP Paribas, Research Division Chris Gao – CLSA Limited, Research Division Maria Meita – Bernstein Institutional Services LLC, Research Division Daria Nasledysheva – BofA Securities, Research Division
Presentation
Advertisement
Operator
Good afternoon, good morning, everyone. Thank you for joining the Ermenegildo Zegna Group First Half 2026 Preliminary Revenues Earnings Call. Please note that today’s material and presentation are available under the zegnagroup.com website.
Before we begin, we need to point out that the team will make certain forward-looking statements during the call. The group’s actual results may be materially different from those expressed or implied by those forward-looking statements. Also, these statements are subject to a number of risks and uncertainties, including those described in our SEC filings. Please refer to the forward-looking statement cautionary statement included at Page 2 of today’s presentation.
I’ll now hand over to Paola Durante, Chief of External Relations and Sustainability.
Advertisement
Paola Durante Chief of External Relations
Thank you. Thank you, operator, and good morning, good afternoon, everyone, and welcome to today’s call. As usual, Gianluca Tagliabue, Group CEO, will share the call, while I will begin with a brief comment on our second quarter revenue results before handing over to Gianluca for some final comments on key events of the quarter that you can see also highlighted in the opening page of the presentation and for some closing remarks.
I remind you that, as always, when commenting on revenue trends, we will focus on organic performance, which excludes foreign
Mark Pownall is joined by Gary Adshead, Nadia Budihardjo and Claire Tyrrell to talk about the big events of the week in WA business and politics. This week the discussion includes Pauline Hanson, Labor’s industrial court, uranium, Andrew Forrest’s tungsten move, Wesfarmers, St Martins redevelopment, Vivian Yap, Greg Poland, Woodside and Aspermont.
OpenAI began rolling out a dedicated health feature to all adult users in the United States on Wednesday, allowing ChatGPT to securely connect to Apple Health data and electronic medical records in an effort to give users more personalized, context-aware answers to health-related questions.
The feature, called Health in ChatGPT, is now available to logged-in users 18 and older across web and iOS platforms, spanning all of the company’s subscription tiers, including Free, Go, Plus and Pro. Users can access the tool by opening Health from the sidebar within the main ChatGPT interface.
What the feature does
Health in ChatGPT allows users to securely link data from Apple Health, along with medical records from supported healthcare systems including Epic and Oracle Health, directly into their conversations with the chatbot. According to OpenAI, the feature can help users compare lab test results against prior readings, track how sleep and activity patterns change over time, monitor medication histories, and prepare for upcoming medical appointments.
Advertisement
“Health in ChatGPT now lets eligible U.S. users securely connect medical records and Apple Health to get more personalized insights and better understand their health,” OpenAI said in a blog post announcing the launch. On social media, the company reiterated the rollout in similar terms, posting, “Health in ChatGPT is starting to roll out to U.S. users. You can securely connect Apple Health and supported medical records to understand your information in context, track what has changed, and have more informed conversations.”
Electronic health record access is powered through a partnership with b.well, a company that aggregates data from roughly 2.2 million U.S. healthcare providers, allowing users to pull in information from a broad range of hospital and clinic systems rather than being limited to a single provider’s patient portal.
Why OpenAI built it this way
OpenAI said the decision to weave health context directly into the main chat interface, rather than isolating it to a separate tool entirely, stemmed from user behavior data gathered during testing. The company found that more than 70% of health-related queries happened organically within the course of everyday conversations, such as a user checking for food allergies while planning a meal, rather than through a dedicated, standalone health tab. OpenAI said the feature nonetheless appears within ChatGPT’s sidebar as its own space with separate chat history and “memories,” giving users the ability to manage their connected health context distinctly from other conversations.
Advertisement
OpenAI emphasized that more than 300 million people use ChatGPT to ask health-related questions on a weekly basis, but said that the underlying context behind those questions is often scattered across separate patient portals, medical records, fitness apps and wearable devices, making it difficult for users to see a complete picture of their health without manually piecing information together themselves.
The models behind the feature
The health feature is powered by two of OpenAI’s most recent models: GPT-5.5 Instant, available to users on the free tier, and GPT-5.6 Sol, offered to paid subscribers. OpenAI said GPT-5.5 Instant has shown meaningful improvement in recognizing when a user’s symptoms may warrant urgent medical attention, asking relevant follow-up questions, and explaining uncertainty in its responses. The company said that model performed at a level comparable to its more advanced “Thinking” models on its most challenging internal health evaluations at the time those evaluations were conducted. GPT-5.6 Sol, meanwhile, is described by OpenAI as its strongest model yet for health-related conversations.
OpenAI said it developed the feature in collaboration with more than 260 physicians practicing across 60 countries and dozens of medical specialties, using their input to help benchmark model performance against real-world clinical scenarios. The company has stressed that the tool is intended to help users feel more informed navigating everyday health questions rather than to serve as a diagnostic service or replacement for professional medical care.
Advertisement
Privacy safeguards
To address privacy concerns tied to handling sensitive medical data, OpenAI said it has implemented several layered safeguards, including strict non-training policies. According to the company, connected medical records, Apple Health data and any chat conversations that draw on that information will not be used to train OpenAI’s foundation models or to inform advertising. By default, the company said, ChatGPT will ask for a user’s permission before referencing connected medical records to generate a given response, giving users ongoing control over when and how their health data is used within the app.
A relaunch, not a first attempt
Wednesday’s broad rollout marks a relaunch of a feature OpenAI first piloted in a more limited form in January 2026. That earlier test period, according to reporting from 9to5Mac, produced what the outlet described as “lackluster results,” prompting OpenAI to spend the following months rebuilding the feature based on user feedback and improvements made to its underlying models since the start of the year. This week’s launch significantly expands access beyond that initial group of testers to include all eligible adult users across the United States.
Advertisement
Context around ongoing scrutiny
The launch comes as OpenAI faces at least one lawsuit related to health-related use of its chatbot. According to reporting, the company was sued by Scott Winters, a 55-year-old former pastor from Florida, who alleged that ChatGPT provided him with inaccurate health advice that contributed to a delay in receiving treatment for a life-threatening pulmonary embolism. OpenAI has not issued a specific public statement addressing that lawsuit in connection with this week’s Health feature launch, and the company continues to describe the tool as intended for informational purposes rather than as a substitute for professional medical evaluation and care.
With Health in ChatGPT now available broadly across the United States, OpenAI is likely to continue refining the feature based on user feedback, following the same pattern that shaped its return after January’s limited pilot. The company has not indicated a timeline for expanding the feature to international markets or additional platforms, noting specifically that Health is not yet available within Codex, OpenAI’s coding-focused product line. For now, the rollout represents one of OpenAI’s most significant pushes yet into integrating personal health data directly into everyday consumer AI conversations, a move likely to draw continued attention from both users and health care privacy advocates in the weeks ahead.
The financial watchdog has banned two Perth advisors for three years after losing a Federal court case regarding an unpaid Australian Financial Complaints Authority determination.
Japan’s Nikkei share average fell more than 2% on Friday, as a sharp decline in Google parent Alphabet shares spurred concerns about heavy AI spending.
The Nikkei was down 2.69% at 64,634.04 as of 0112 GMT, while the broader Topix slipped 1.28% to 4,002.09.
The Nikkei has lost more than 7% so far this month, tumbling into correction territory last week. Its moves have been heavily affected by the tech-heavy South Korean benchmark KOSPI and the U.S. Philadelphia semiconductor index.
Shares of Alphabet sank 7% overnight after the company reported higher spending plans while it also burned cash. Wall Street indexes closed lower, with the Nasdaq shedding more than 2%.
Advertisement
Concerns resurfaced over whether heavy spending on AI infrastructure is sustainable after Alphabet shares fell sharply overnight, said Kazuaki Shimada, chief strategist at IwaiCosmo Securities.
Live Events
“The (Nikkei) index has been affected by overseas factors, not local cues. Many Japanese companies will start reporting their earnings from today, and if their outlook is strong, the index’s trend may change,” said Shimada. Chip-related shares fell, with Advantest and Tokyo Electron losing 6.33% and 5.43%, respectively. Technology investor SoftBank Group fell 7.42% and memory chip maker Kioxia lost 4.4%.
Shares supported by domestic demand rose, with Central Japan Railway and East Japan Railway rising 1.17% and 0.6%, respectively.
Shippers rose, with Kawasaki Kisen and Mitsui OSK Lines up 0.61% and 0.88%, respectively.
Otsuka Holdings, a maker of Pocari Sweat, rose 1.6% to become the top percentage gainer on the Nikkei.
China’s export machine produced one of the most consequential economic statistics of recent years when it posted a record trade surplus of approximately USD 1.2 trillion for 2025. China’s customs data revealed exports hit USD 3.77 trillion in 2025, up 5.5% year-on-year, while imports remained flat at USD 2.58 trillion, yielding an unprecedented surplus that is equivalent to the GDP of a top-20 economy.
For Thailand — China’s largest trading partner in ASEAN and one of the region’s most export-dependent economies — the implications of that figure are neither straightforwardly good nor unambiguously bad. They are, more precisely, a study in structural tension: a country that benefits from Chinese investment and supply chain integration while simultaneously absorbing the competitive pressure of Chinese overcapacity across sector after sector.
Understanding that tension is not an academic exercise. It is the operating reality for Thai manufacturers, retailers, policymakers, and investors navigating 2026.
The Trade Deficit That Keeps Widening
The bilateral trade picture tells its own story. In 2025, Thailand’s exports to China reached USD 39.72 billion, while imports from China were significantly higher at USD 107.62 billion — a trade deficit that has grown every single year for the past five years.
Advertisement
Thailand’s deficit with China rose 50% in 2025 to USD 67.8 billion — among the steepest annual widening in the region, alongside Malaysia’s 62% jump and Vietnam’s 40% increase. Analysts point to a direct connection: as China has faced elevated US tariffs, its goods have increasingly been redirected toward Southeast Asian markets, raising import volumes across the board.
The contrast with the US trade relationship is stark. Thailand’s total trade with the US reached USD 93.65 billion in 2025, generating a Thai surplus of THB 51.4 billion. Electronics and telecommunications products drove that outperformance, with exports to the US generating a positive trade balance every year of the past five.
Thailand is therefore caught in an increasingly uncomfortable asymmetry: earning from the West what it owes to the East — a structural position that creates both opportunity and vulnerability as the US-China trade war intensifies.
The Factory Closure Crisis
The most visible domestic consequence of China’s export surge is industrial contraction. Thailand’s Kasikorn Research Center estimates that 4,300 Thai factories closed in the two years to 2025, spanning furniture, electronics, garments, automotive, and steel — with the trend expected to worsen.
The automotive sector illustrates the dynamic with particular clarity. Suzuki Motor Thailand announced the closure of its Thai production plant by the end of 2025 due to declining sales and competition from Chinese electric vehicles. Tan Chong Subaru Automotive Thailand ceased car production by 2024 due to ongoing losses, affecting parts suppliers across the ICE component supply chain.
Advertisement
Steel is another pressure point. Capacity utilisation in Thailand’s steel industry fell to approximately 29.3% in early 2024, down from 32.4% the year before. The anti-dumping response has been measured but assertive: Thailand has imposed duties of 31% on hot-rolled coil from China, covering high-strength steel used in critical infrastructure. The Anti-Dumping and Subsidy Review Committee has agreed to additional retaliatory measures targeting 33 product categories, running from October 2025 to October 2030.
The scope of Chinese overcapacity has also expanded significantly. Unlike earlier periods when Chinese excess production was largely restricted to commodities like textiles, steel, and aluminium, overcapacity now extends into food processing, pharmaceuticals, and certain chemical products — products far closer to Thailand’s higher-value manufacturing base.
The EV Paradox: Opportunity and Disruption Simultaneously
No sector better captures the dual nature of China’s economic presence in Thailand than electric vehicles. Thailand has positioned itself as the EV hub of Southeast Asia — and Chinese manufacturers have been central to building that ambition. Yet those same manufacturers are simultaneously displacing the Japanese automotive industry that formed the bedrock of Thailand’s manufacturing economy for four decades.
Chinese automakers held an 82% market share in Thailand’s battery electric vehicle segment as of 2024, with BYD commanding 40% and Hozon Auto and SAIC Motor controlling a combined 35%. Facing that concentration of Chinese and Korean competition, Japanese firms managed to capture less than 1% of the EV market.
Advertisement
By 2026, Chinese companies plan to produce more than one million vehicles in ASEAN countries, with approximately 600,000 expected to be EVs — more than half of China’s entire overseas production capacity. Thailand and the Philippines ranked among China’s largest EV export markets in 2024.
This creates a policy dilemma with no clean resolution. Welcoming Chinese EV investment builds the next-generation automotive sector that Thailand needs. But it also accelerates the hollowing out of the Japanese-anchored supply chains that currently employ hundreds of thousands of Thai workers. The government’s ability to manage that transition — through local content requirements, supplier development programs, and workforce reskilling — will define the long-term terms of the bargain.
The Transshipment Risk
Complicating Thailand’s position further is the growing scrutiny of transshipment — the practice of routing Chinese goods through third countries to avoid US tariffs.
Thailand faces a surge in imports of goods linked to transshipment: items rerouted to bypass US tariffs, or products falsely declared as originating elsewhere. The US has imposed anti-dumping and countervailing duties on solar cells imported from Thailand, Malaysia, Vietnam, and Cambodia, effective April 2025 — with rates set at 375% to 972% for Thai manufacturers. The scale of those duties signals the severity of Washington’s concern and represents a direct threat to Thai solar exporters who are not transshipping but are caught in the regulatory blowback nonetheless.
Thailand’s import content — particularly from China — has risen significantly in recent years, constraining the domestic benefits of incoming investment and increasing the risk of additional US transshipment tariffs going forward.
Advertisement
The Macro Outlook: The Weakest Growth in a Generation
The cumulative effect of these pressures is registering in Thailand’s growth trajectory. The SCB Economic Intelligence Center projects Thailand’s economy to expand by only 1.5% in 2026, down from 2% in 2025 — the lowest growth in three decades outside of crisis periods. The IMF has issued a forecast of 1.6% and the World Bank anticipates approximately 1.7%.
The export sector, which accounts for a substantial share of Thailand’s GDP, faces multiple headwinds: the fading effect of front-loading ahead of US tariffs, rising risks of additional duties on electronics and transshipped products, and intensifying competition from China following its one-year trade agreement with the US to temporarily reduce retaliatory tariffs.
The trade war and slowing external demand from the US will pose risks to Thailand’s manufacturing output and export growth, either directly or indirectly through Thai manufacturers linked to Chinese firms. The Bank of Thailand has cut rates to 1.25% and further easing is expected in 2026, but monetary policy alone cannot address structural competitiveness gaps.
The Policy Response: Necessary but Incomplete
Prime Minister Paetongtarn Shinawatra’s administration has signalled intentions to review import duties and promote local content requirements to shield industries. The NESDC has urged vigilance, noting that while China’s surplus highlights Beijing’s economic resilience, it also underscores the need for ASEAN unity to address trade imbalances.
Advertisement
Experts argue that Thailand’s response must go beyond reactive tariff measures. Bolstering domestic manufacturing through incentives, investing in high-value sectors like electric vehicles, and negotiating fairer trade terms with China are identified as the core pillars of a sustainable strategy. The challenge is sequencing those ambitions against an immediate industrial contraction that is moving faster than policy frameworks can adapt.
China’s record trade surplus is not a problem that Thailand can opt out of. The two economies are too deeply integrated — in supply chains, in investment flows, in tourism, and in digital infrastructure — for Bangkok to meaningfully decouple. What Thailand can do is manage the terms of that integration more deliberately: directing Chinese investment toward sectors that build long-term industrial capability, reinforcing anti-dumping mechanisms against predatory pricing, developing domestic supply chains resilient enough to withstand competitive displacement, and positioning clearly enough in the US trade relationship to avoid the transshipment penalties that would compound an already difficult export environment.
The mixed fortunes embedded in China’s trade surplus are not going away. How Thailand navigates them over the next two to three years will do much to determine whether the country emerges from this period of economic stress with its industrial base intact — or significantly diminished.
Sources: Thailand Business News; Nation Thailand; Kasikorn Research Center; SCB Economic Intelligence Center; Asia Society Policy Institute; Krungsri Research Industry Outlook 2026–2028; Allianz Trade Country Report Thailand; East Asia Institute
You must be logged in to post a comment Login