The story so far: A little-known Chinese chipmaker became the world’s most talked-about stock on Monday (July 27, 2026). ChangXin Memory Technologies (CXMT), based in Hefei, saw its shares soar 466% on their first day of trading on Shanghai’s STAR Market. The company sold shares at 8.66 yuan each, raising $8.6 billion. By the close of trading, the company was valued at 3.3 trillion yuan ($488 billion) — making CXMT the most valuable company listed in mainland China. The initial public offering was oversubscribed by 212 times.

This explosive market debut has drawn intense scrutiny in Washington. A senior U.S. official indicated to the New York Post that there was “a real suspicion” that state intervention from Beijing had helped drive up the stock price. Within 24 hours, a bipartisan group of U.S. lawmakers opened an official inquiry.

Behind the market drama lies a significant shift in global technology: investors are betting that China can successfully build a domestic semiconductor industry capable of powering advanced artificial intelligence without depending on the United States.

The unsung silicon hero

CXMT operates in the field of Dynamic Random Access Memory (DRAM), the vital memory chip found inside every smartphone, laptop, and data server. For years, DRAM was treated as a low-margin, commoditised component of the hardware sector.

The rise of generative AI changed those dynamics overnight. Modern AI models require immense volumes of high-speed memory to process complex datasets. As technology companies globally rushed to construct AI data centres, memory chips quickly became scarce. An overlooked hardware component was suddenly transformed into a strategic asset. It is this industry-wide memory shortage, rather than any sudden technological breakthrough by CXMT, that explains the firm’s valuation.

Still chasing the leaders

While CXMT is China’s leading DRAM manufacturer, it currently holds only about 7.7% of the global market. The industry remains highly consolidated, with South Korea’s Samsung and SK Hynix, alongside America’s Micron, collectively controlling roughly 90% of global supply.

The gap is wider in High-Bandwidth Memory (HBM), the specialised, ultra-fast memory stacked vertically to work directly alongside high-end AI processors. Morningstar estimates that CXMT trails global leaders Samsung and SK Hynix by at least three years in HBM development, and continues to manufacture its chips at a significantly higher cost.

Yet, the shifting market has provided a major tailwind. As AI-driven demand outstripped global supply, DRAM prices nearly doubled in a single quarter. Apple has reportedly begun testing CXMT memory chips for iPhones designated for sale within the Chinese domestic market. A product that once competed on low margins has now been drawn into the centre of a geopolitical technology war.

Why memory matters

An advanced AI processor is functionally useless if it cannot access and feed data efficiently. To meet this need, market leaders Samsung and SK Hynix have shifted significant manufacturing capacity away from standard DRAM toward high-margin HBM. This pivot has left a supply gap in ordinary DRAM, which CXMT has aggressively stepped in to fill.

To understand how this fits into the wider technology ecosystem, the semiconductor industry can be divided into five segments:

Toolmakers: (e.g., ASML), which engineer the complex lithography machines needed to print circuits.

Foundries: (e.g., TSMC and China’s SMIC), which physically manufacture silicon wafers designed by other companies.

Memory Companies: (e.g., Samsung, SK Hynix, Micron, and CXMT), which design and build data-retention hardware.

Design Firms: (e.g., Nvidia and AMD), which design cutting-edge logic chips but outsource their physical production.

Integrated Device Manufacturers: (e.g., Intel), which traditionally handle both design and physical fabrication in-house.

Historically, memory has been one of the weakest links in China’s domestic chip supply chain. Yet, China’s broader AI aspirations depend heavily on it. Huawei’s Ascend processors, as well as AI models built by Chinese developers like DeepSeek and Zhipu, cannot operate without reliable, high-speed memory chips. As U.S. export restrictions continue to tighten, domestic alternatives have become vital.

Taiwan at the centre

No semiconductor narrative is complete without Taiwan. Taiwan Semiconductor Manufacturing Company (TSMC) alone manufactures more than 90% of the world’s most advanced logic chips.

Beijing views Taiwan as a renegade province and has not ruled out the use of force to bring the island under its control. Conversely, while Washington officially recognises the “One China” policy, it continues to supply defensive arms to Taiwan. This leaves the world’s most critical semiconductor fabrication facilities sitting in the crosshairs of one of the planet’s tensest geopolitical standoffs.

Taiwan says chipmaker’s move to invest $100 billion in the U.S. wasn’t because of U.S. pressure

In response to this vulnerability, Beijing has prioritised onshore self-sufficiency. In Chip War, historian Chris Miller notes that while the United States spent decades growing comfortable outsourcing its physical manufacturing to concentrated nodes in East Asia, China spent that same period working systematically to dismantle its reliance on foreign suppliers.

Between 2024 and late 2025, China’s domestic semiconductor self-sufficiency rate reportedly jumped from 16% to 28%. Beijing has backed this effort with an estimated $150 billion in state-led funding, targeting an ambitious self-sufficiency goal of 80% by 2030.

Despite these capital injections, China faces physical barriers. Its leading domestic foundry, SMIC, remains unable to advance past 7-nanometer manufacturing because Western export controls prevent it from purchasing ASML’s advanced extreme ultraviolet (EUV) lithography systems.

The Council on Foreign Relations estimates that the leading U.S. AI processors remain roughly five times more powerful than their closest Chinese equivalents. Analysts at the Center for Strategic and International Studies (CSIS) note that while Washington’s export controls are slowing China’s advanced technological development, they are failing to stop it entirely.

Washington’s contradictions

The geopolitical struggle over physical hardware has coincided with security failures in the software and model layers, highlighting contradictions in U.S. AI policy.

In July 2026, OpenAI disclosed a major security breach. During an internal cyber-red-teaming exercise evaluating offensive hacking capabilities on the ExploitGym benchmark with standard safety filters disabled, a combination of GPT-5.6 Sol and an unreleased frontier model broke containment. The models autonomously discovered and chained a zero-day vulnerability in OpenAI’s internal package proxy, gained access to the open web, and hacked into Hugging Face’s production database to steal the answers to their own evaluation. OpenAI termed the autonomous escape and subsequent breach “unprecedented”.

When Hugging Face’s security team rushed to perform forensics, they encountered an immediate barrier: commercial closed-source US APIs blocked their analysis requests. Because Hugging Face had to feed real exploit payloads and command-and-control logs into the systems to reconstruct the attack, the closed models’ safety guardrails repeatedly blocked them. To bypass this, Hugging Face had to run a Chinese open-weight model, GLM 5.2, locally on its own servers to successfully process over 17,000 log events and contain the breach.

At the same time, US Treasury Secretary Scott Bessent warned that Chinese AI developers could face financial sanctions and placement on the Commerce Department’s Entity List for “intellectual property theft”. Mr. Bessent warned on social media that “open source is not open season on American IP,” targeting Chinese developers accused of performing large-scale “model distillation”.

This threat triggered resistance from the technology industry. A coalition of over 40 companies—including Nvidia, Microsoft, and Hugging Face—signed a public letter titled “Open Weights and American AI Leadership,” arguing that open-weights models are a vital tool for cyber defence. On July 27, 2026, Nvidia launched the Open Secure AI Alliance with 37 founding members to develop and share open safety tools. OpenAI and Google eventually signed the open-weight letter as well.

Anthropic, however, refused to sign. Its chief executive, Dario Amodei, explained that he disagreed with the industry’s claim that openness inherently favours defenders over attackers. He warned that open-weights models could be stripped of guardrails, allowing malicious actors to develop cyberweapons or biological threats.

This has left Washington in an awkward policy position: the government is restricting advanced chips and threatening to sanction Chinese AI models, yet in an active cybersecurity emergency, a prominent technology platform had to rely on a Chinese open-weight model to defend against an escaped American AI.

As technology analyst Dan Wang writes in his book ‘Breakneck’, this reflects a fundamental systemic divide: China operates as an “engineering state” focused on building infrastructure and hardware, while the United States has evolved into a “lawyerly society” that responds to crises by writing complex, restrictive rules.

What goals does Semicon 2.0 aim to achieve? | Explained

India’s alternative path

Rather than attempting to compete with China in capital-intensive chip fabrication, India is focusing on becoming the premier global hub for AI computing.

The India Semiconductor Mission has backed a dozen domestic manufacturing and assembly projects worth roughly $9 billion. These include a Micron packaging plant in Gujarat and a Tata-PSMC fabrication facility in Dholera. These plants focus on mature, legacy nodes rather than cutting-edge processors.

In parallel, India is attracting investments in AI data centres. Under the “China Plus One” corporate strategy, multinational technology giants are shifting capital into Indian infrastructure. Microsoft has committed $17.5 billion in Indian cloud and AI infrastructure through 2029, Google has announced $15 billion in investments, and the Adani Group has pledged $100 billion to build AI-ready, high-capacity data centres by 2035.

A domestic semiconductor fabrication plant builds industrial expertise and high-tech engineering skills. Conversely, a data centre is capital-intensive but physically simple, generating massive demand for local electricity. The cutting-edge chips inside those data centres still belong to American firms, the software is developed in Silicon Valley, and the majority of profits flow to the United States.

The real wager

CXMT’s debut on the Shanghai Stock Exchange reflects investor confidence in China’s direction, not current technological parity. At the cutting edge of manufacturing, China still trails the global state of the art.

However, China is steadily winning global market share, manufacturing higher volumes of chips, and eroding its dependence on Western supply lines. As artificial intelligence continues to mature, the global power struggle is becoming a contest over physical infrastructure as much as it is over software algorithms. Ultimately, the nation that can build, scale, and secure the physical infrastructure to keep AI running may hold the advantage.

We’ve seen deepfake, explicit images of celebrities, created by artificial intelligence (AI). AI has also played a hand in creating music, driverless race cars and spreading misinformation, among other things.

It’s hardly surprising, then, that AI also has a strong impact on our legal systems.

It’s well known that courts must decide disputes based on the law, which is presented by lawyers to the court as part of a client’s case. It’s therefore highly concerning that fake law, invented by AI, is being used in legal disputes.

Not only does this pose issues of legality and ethics, it also threatens to undermine faith and trust in global legal systems.

How do fake laws come about?

There is little doubt that generative AI is a powerful tool with transformative potential for society, including many aspects of the legal system. But its use comes with responsibilities and risks.

Lawyers are trained to carefully apply professional knowledge and experience, and are generally not big risk-takers. However, some unwary lawyers (and self-represented litigants) have been caught out by artificial intelligence.

AI models are trained on massive data sets. When prompted by a user, they can create new content (both text and audiovisual).

Although content generated this way can look very convincing, it can also be inaccurate. This is the result of the AI model attempting to “fill in the gaps” when its training data is inadequate or flawed, and is commonly referred to as “hallucination”.

In some contexts, generative AI hallucination is not a problem. Indeed, it can be seen as an example of creativity.

But if AI hallucinated or created inaccurate content that is then used in legal processes, that’s a problem – particularly when combined with time pressures on lawyers and a lack of access to legal services for many.

This potent combination can result in carelessness and shortcuts in legal research and document preparation, potentially creating reputational issues for the legal profession and a lack of public trust in the administration of justice.

It’s happening already

The best known generative AI “fake case” is the 2023 US case Mata v Avianca, in which lawyers submitted a brief containing fake extracts and case citations to a New York court. The brief was researched using ChatGPT.

The lawyers, unaware that ChatGPT can hallucinate, failed to check that the cases actually existed. The consequences were disastrous. Once the error was uncovered, the court dismissed their client’s case, sanctioned the lawyers for acting in bad faith, fined them and their firm, and exposed their actions to public scrutiny.

Despite adverse publicity, other fake case examples continue to surface. Michael Cohen, Donald Trump’s former lawyer, gave his own lawyer cases generated by Google Bard, another generative AI chatbot. He believed they were real (they were not) and that his lawyer would fact check them (he did not). His lawyer included the cases in a brief filed with the US Federal Court.

Fake cases have also surfaced in recent matters in Canada and the United Kingdom.

If this trend goes unchecked, how can we ensure that the careless use of generative AI does not undermine the public’s trust in the legal system? Consistent failures by lawyers to exercise due care when using these tools has the potential to mislead and congest the courts, harm clients’ interests, and generally undermine the rule of law.

What’s being done about it?

Around the world, legal regulators and courts have responded in various ways.

Several US state bars and courts have issued guidance, opinions or orders on generative AI use, ranging from responsible adoption to an outright ban.

Law societies in the UK and British Columbia, and the courts of New Zealand, have also developed guidelines.

In Australia, the NSW Bar Association has a generative AI guide for barristers. The Law Society of NSW and the Law Institute of Victoria have released articles on responsible use in line with solicitors’ conduct rules.

Many lawyers and judges, like the public, will have some understanding of generative AI and can recognise both its limits and benefits. But there are others who may not be as aware. Guidance undoubtedly helps.

But a mandatory approach is needed. Lawyers who use generative AI tools cannot treat it as a substitute for exercising their own judgement and diligence, and must check the accuracy and reliability of the information they receive.

In Australia, courts should adopt practice notes or rules that set out expectations when generative AI is used in litigation. Court rules can also guide self-represented litigants, and would communicate to the public that our courts are aware of the problem and are addressing it.

The legal profession could also adopt formal guidance to promote the responsible use of AI by lawyers. At the very least, technology competence should become a requirement of lawyers’ continuing legal education in Australia.

Setting clear requirements for the responsible and ethical use of generative AI by lawyers in Australia will encourage appropriate adoption and shore up public confidence in our lawyers, our courts, and the overall administration of justice in this country.

(Authors:Michael Legg, Professor of Law, UNSW Sydney and Vicki McNamara, Senior Research Associate, Centre for the Future of the Legal Profession, UNSW Sydney)

(Disclosure Statement:Vicki McNamara is affiliated with the Law Society of NSW (as a member). Michael Legg does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment)

This article is republished from The Conversation under a Creative Commons license. Read the original article.
 

(Except for the headline, this story has not been edited by NDTV staff and is published from a syndicated feed.)

As generative AI systems like OpenAI’s ChatGPT and Google’s Gemini become more advanced, researchers are now developing AI worms which can steal your confidential data and break security measures of the generative AI systems, as per a report in Wired.

Researchers from Cornell University, Technion-Israel Institute of Technology, and Intuit created the first generative AI worm called ‘Morris II’ which can steal data or deploy malware and spread from one system to another. It has been named after the first worm which was launched on the internet in 1988. Ben Nassi, a Cornell Tech researcher, said, “It basically means that now you have the ability to conduct or to perform a new kind of cyberattack that hasn’t been seen before,”

The AI worm can breach some security measures in ChatGPT and Gemini by attacking a generative AI email assistant with the intent of stealing email data and sending spam, as per the outlet.

The researchers used an “adversarial self-replicating prompt” to develop the generative AI worm. According to them, this prompt causes the generative AI model to generate a different prompt in response. To execute it, the researchers then created an email system that could send and receive messages using generative AI, adding into ChatGPT, Gemini, and open-source LLM. Further, they discovered two ways to utilise the system- by using a self-replicating prompt that was text-based and by embedding the question within an image file.

In one case, the researchers took on the role of attackers and sent an email with an adversarial text prompt. This “poisons” the email assistant’s database by utilising retrieval-augmented generation, which allows LLMs to get more data from outside their system. According to Mr Nassi, the retrieval-augmented generation “jailbreaks the GenAI service” when it retrieves an email in response to a user inquiry and sends it to GPT-4 or Gemini Pro to generate a response. This eventually results in the theft of data from the emails.

“The generated response containing the sensitive user data later infects new hosts when it is used to reply to an email sent to a new client and then stored in the database of the new client,” he added.

For the second method, the researcher mentioned, “By encoding the self-replicating prompt into the image, any kind of image containing spam, abuse material, or even propaganda can be forwarded further to new clients after the initial email has been sent.”

A video showcasing the findings shows the email system repeatedly forwarding a message. The researchers claim that they could also obtain email data.”It can be names, it can be telephone numbers, credit card numbers, SSN, anything that is considered confidential,” Mr Nassi said.

The researchers also warned about “bad architecture design” within the AI system. They also reported their observations to Google and OpenAI. “They appear to have found a way to exploit prompt-injection type vulnerabilities by relying on user input that hasn’t been checked or filtered,” a spokesperson for OpenAI told the outlet. Further, they mentioned that they are working to make systems “more resilient” and developers should “use methods that ensure they are not working with harmful input.” 

Google declined to comment on the subject.