As the election day gets closer, the volume from the rumour mill and smear campaigns is clearly being turned up.
Among the latest claims doing the rounds is a story that a company is paying HK$150 per head to “hire voters” for the Legislative Council election on December 7, and even demanding they take photos as proof after casting their ballots, which has triggered a wave of anti-government comments online.
In response, the Office for Safeguarding National Security of the CPG in the HKSAR has issued its third statement on the matter, stressing that certain people inside and outside Hong Kong keep changing tactics in order to smear the Legislative Council election.
A spokesperson for the Office pointed out that some anti‑China, destablising-Hong Kong elements and internal forces have not abandoned their malicious intentions or wicked ambitions, and are still openly and covertly inciting people to boycott the election and stirring up disruptive activities both online and offline.
The Office has made clear it will take a zero‑tolerance approach in defending national security and, in accordance with the law, will severely punish any acts or activities that interfere with or sabotage the election and endanger national security.
Looking carefully at the online posts, it is obvious that the National Security Office’s statement hits the target.
Take the above‑mentioned post about so‑called “paid voting”, offering HK$150 per person to get people to vote and then take a photo as evidence – its source was unclear and the whole setup was already rather bizarre from the start.
Twisting “Paid Voting” Into a Weapon
When the media asked the Independent Commission Against Corruption (ICAC) about this, its spokesperson explained that the activity was purely to promote the election, without inducing voters to vote or not vote for particular candidates, and therefore it is not restricted by the Elections (Corrupt and Illegal Conduct) Ordinance.
At the same time, the Office of the Privacy Commissioner for Personal Data reminded the public to first find out the purpose of any data collection and to pay attention to whether the related information is genuine.
Yet on the forums where this piece of news was reposted, some comments seized the chance to add fuel to the fire and attack the government, while others simply fabricated rumours.
One example was a comment claiming that “According to Section 11 of the Elections (Corrupt and Illegal Conduct) Ordinance, any person who, without reasonable excuse, offers a benefit to another person to induce or reward that person or a third party to vote or not to vote in an election commits an offence.”
As soon as this post appeared, it encouraged many people to rush out and condemn the government and the ICAC, arguing that since the ordinance “clearly” bans offering benefits to induce others to vote, how could the ICAC possibly say that encouraging voting is not restricted by the law.
However, simply checking Section 11 of the Elections (Corrupt and Illegal Conduct) Ordinance is enough to show that its content is completely different from what the commenter wrote.
Section 11 targets corrupt conduct involving bribery of voters or others in an election, and makes it an offence for any person, without reasonable excuse, to offer a benefit to another person as an inducement for that person to vote for a particular candidate.
When Fake Law Becomes a Political Tool
In reality, Section 11 is about “offers an advantage to another person as an inducement to vote at the election for a particular candidate or particular candidates”, not “offering a benefit to another person to vote” in general.
The commenter deliberately rewrote the law, changing its original ban on offering benefits to get people “to vote for a particular candidate” into a supposed ban on offering benefits just to get people “to vote”, which is an obvious attempt to manufacture a rumour.
That comment then triggered a flood of insults against the ICAC.
In fact, this kind of deliberate fabrication of legal content is already suspected of breaching the offence of seditious intention under Article 24 of the Safeguarding National Security Ordinance.
Article 24 makes clear that a seditious intention includes the intention to incite hatred or contempt against the executive, legislative or judicial authorities of the HKSAR.
By inventing the content of the ordinance and then accusing the ICAC of failing to enforce the law, the commenter stirs up hatred against the ICAC and is therefore suspected of committing the offence of seditious intention.
Are the flood of specious messages appearing online, the doctored legal provisions and the calls to hate the government and to refuse to vote on December 7 just casual off‑the‑cuff remarks, or are they deliberate acts of incitement?
Don’t Play With Legal Fire Online
Even if an ordinary person wanted to criticise the government, it is unlikely they would spend the time and effort to rewrite the content of the Elections (Corrupt and Illegal Conduct) Ordinance, which makes this behaviour particularly suspicious.
Now that the National Security Office has spoken out in advance and warned that it will, in accordance with the law, severely punish all acts intended to interfere with or sabotage the election, this already serves as a serious and timely warning.
These individuals are intentionally spreading false information aimed squarely at the election, and it is highly likely that the posters or commenters are not physically in Hong Kong.
They maliciously attack the city’s system while confidently assuming the government will find it hard to hold them to account.
The real problem is that if any people in Hong Kong foolishly forward such content online, and even call on others to attack the government or boycott the election by refusing to vote, they can very easily end up breaking the law.
Clearly, attempts to attack Hong Kong’s election are no simple matter, and others should not casually chime in and “test the law by themselves”.
Lo Wing-hung
Bastille Commentary
** 博客文章文責自負,不代表本公司立場 **
China's strategy is simple: throw one punch to avoid a hundred.
On Wednesday, August 5, China fired back with five consecutive moves to counter a string of recent US measures that Beijing sees as hostile. A spokesperson for China's Ministry of Commerce called the countermeasures "generally restrained." The message was clear: since the China-US presidential meeting in Busan, the US Federal Communications Commission has ignored China's strong opposition and industry appeals. It keeps stretching the concept of national security to roll out restrictions on China.
Meanwhile, the US has added over 40 Chinese entities to the so-called Uyghur Forced Labor Prevention Act entity list. China demands the US immediately revoke these measures. And if Washington insists on new restrictions, "China will further retaliate."
China's moves are a direct answer to a series of petty US actions that fall into two main areas.
First, the FCC. Before the government took action, the People's Daily fired a warning shot. Under the byline "Zhong Sheng," it blasted the FCC for hiding behind banners like "non-discrimination" and "national security" to discriminate against and suppress Chinese companies.
The FCC recently added foreign-made power inverters and advanced robotic equipment to its so-called "Covered List." That means new models can't get certification and are locked out of the US market. The paper called the tactic typical unilateral bullying.
This isn't new. Back in 2021, the FCC put telecom and video surveillance gear from five Chinese companies on the list. Since then, it has steadily widened the net. In April, it proposed revoking the qualifications of testing and certification bodies from countries that haven't signed a "Mutual Recognition Agreement" with the US. That artificially raises the compliance bar for Chinese products.
And now, Reuters reports the FCC is drafting a ban on imports of new Chinese-made optical communication transceivers for data centers.
Second, the Xinjiang sanctions. On July 30, the US Department of Homeland Security added over 40 Chinese entities to the entity list under the Uyghur Forced Labor Prevention Act. The move, effective August 3, swells the list from 144 to 187, the biggest expansion since it was created. The new targets span food, clothing, and cotton. They include snack maker Qia Qia melon seeds, frozen food producer Zhengzhou Synear dumplings, and apparel brand Fujian Septwolves.
When the news hit the Chinese internet, netizens erupted in mockery. The joke going around teasing that "eating melon seeds can easily damage US soldiers' teeth."
China didn't waste a second. It fired back with five countermeasures, and the first one hits where it hurts: drones. Beijing is tightening export controls on drones and related technologies to the US. From now on, every shipment of drones, key components, and controlled technologies will face a strict case-by-case review.
And forget about any 'licensing facilitation' exemptions—those are off the table. Think about it: China is the world's drone superpower. The US wanted to block Chinese drones from coming in. Now China is flipping the script—it's restricting what goes out. The message is blunt: just because you want to buy doesn't mean we'll sell. Especially not advanced drones and components with military uses.
The second measure is a highly targeted strike: China has added six US entities to its Countermeasures List, all of which it says assisted US sanctions on Xinjiang-related companies.
Applied DNA Sciences, Inc. is believed to provide DNA-based traceability and verification services for relevant products.
Stratum Reservoir, LLC is believed to conduct stable-isotope testing and related analysis to determine the origin of Xinjiang-linked materials.
Altana Technologies, Inc., which operates a global supply-chain mapping and management platform, is believed to help the US government trace supply chains involving Xinjiang-linked products.
The Responsible Business Alliance, which promotes responsible business conduct in global supply chains and operates assessment programmes for member companies, is believed to help companies exclude Xinjiang-linked products from their supply chains.
Verité Group, Inc., a nonprofit focused on labour rights and human rights, examines alleged forced labour in global supply chains.
Human Rights in China, a New York-based human-rights organization, advocates sanctions related to Xinjiang.
China’s countermeasures against these six entities are therefore highly targeted.
The third measure targets a US compliance testing company that helped the FCC impose China-related sanctions. That company will now be locked out of the compliance testing business in the Chinese market. The ripple effect: it could drive up compliance costs for US products trying to enter China.
The fourth measure hits at the heart of product certification. China is suspending the entrustment of US agencies to conduct factory follow-up inspections for CCC certification. Remember, CCC certification is a must-have for US products to enter the Chinese market. By pulling the plug on US inspectors, China shifts the review work to its own agencies. That spells more time and more uncertainty for American exporters.
The fifth measure, a national security investigation into imported US printing and copying office equipment. On the surface, this one looks like a light tap.
China imported about $2.08 billion worth of such equipment in 2025, mostly from Japan ($1.72 billion). The US slice was small. But this is just the opening move.
The US has banned Chinese software and hardware from connected and autonomous vehicles on American roads—and until now, China held its fire.
Now Beijing is investigating US printers and copiers on national security grounds. The message is unmistakable: if Washington plays the national security card at every turn, Beijing can just as easily slap an 'unsafe' label on American phones and EVs running US software.
But that's not the main battlefield. Artificial intelligence is the real arena. Over the past month, Chinese companies have rolled out one high-performance, low-cost AI model after another, sending US AI giants reeling. When Anthropic released a new model in June, the US government demanded it be kept out of non-American hands. The clash forced Anthropic to pull the model. Trump signed an executive order in June setting up an AI safety framework, including a voluntary program for companies to submit cutting-edge models for government review.
Then Chinese developers like Moonshot AI and DeepSeek unleashed open-source systems that rival top-tier US models, fueling a fierce debate in Washington over how to regulate open-weight models. Anthropic CEO Dario Amodei pushed for mandatory government safety reviews for both open and proprietary models, taking direct aim at Chinese open-source systems. He won backing from Treasury Secretary Bessent and others, but ran into a wall of opposition from US tech giants, including Nvidia.
On August 4, the Trump administration delivered its verdict. In a meeting led by the Office of the National Cyber Director, top US AI companies were told that under the new AI safety framework, open-source models from Chinese rivals would not face US government safety testing.
For Anthropic, it was a crushing defeat. US National Cyber Director Sean Cairncross later spelled out the thinking at a cybersecurity conference in Las Vegas: the US government wants to back the American open-source AI ecosystem. He called such models 'tremendously valuable' and insisted the regulatory framework must stay 'flexible.'
Otherwise, he warned, 'it would not only stifle growth, development, and innovation, but the regulatory regime would become obsolete within 48 hours of completing its process.'
The bottom line: Washington keeps chipping away at China with petty moves. But when the pile gets high enough, Beijing punches back hard—and signals it has heavier blows ready.
President Xi is due to visit the US in September, and Trump will be in Shenzhen in mid-November for the APEC summit. Trump doesn't want things to spiral out of control; at crunch time, he'll yank the leash on his officials.
The real battlefield is artificial intelligence. It was almost certain that the White House's decision to leave Chinese open-source models alone came after fierce pushback from China.
Lo Wing-hung
