We also laid out four paths for media companies to navigate the AI era. Path 1 — becoming a data supplier to AI will surely lead to self-disintegration. Path 2 — building your own AI layer is unlikely to be economically viable for most, except the largest media companies. This is similar to Google’s strategy with Gemini: building an AI layer to ensure its search platform is not disintermediated. For most publishers, Path 3 — original journalism — is most rational and is possible, but unlikely on its own to be sustainable. Publishers that also achieved Path 4 — turning its brand into identity — will define success and failure, but only a very few will succeed as the source of trust and truth. Like the thousands of luxury brands, but only a few succeed, as curators of scarcity.
And as we promised in the conclusion of Part 7 last week, this article will articulate that media companies still retain one remaining advantage — their last remaining moat — that can enable them to coexist successfully with AI. Indeed, it shows how AI can be turned from a competitor into a multiplier.
But instead of beginning with a theoretical construct, we will begin with evidence and facts — a real-life recent example. It will help you understand our proposition better.
The example involves the recent news on so-called “corporate mafia” — businessmen using enforcement agencies to do their bidding in Malaysia.
See also: The judgement market: lessons from finance
After reading the articles “Explaining the NexG saga”, “Who’s who in the NexG saga” and “RCI needed to validate procurement process and independence of enforcement agencies” (The Edge, March 30, 2026) on March 28, 2026 (Saturday), I asked ChatGPT: “There are recent news reports in Malaysia about businessmen using mafia-like operations to take over companies. Enforcement agencies are used as tools to enable these corporate players. Please provide context and information.”
The following is ChatGPT’s exact response to my question:
See also: The future of media in the AI age (Part 7/9): The interpretation economy
The “corporate mafia” narrative in Malaysia is not a proven system but serious ongoing allegations sitting in the intersection of business, enforcement power and politics. To understand it properly, you need to separate (a) the claims, (b) the institutional responses and (c) the deeper structural risks.
1. What the allegations actually say.
Recent reports (notably from Bloomberg and picked up locally) described a pattern: A group of businessmen (“corporate mafia”) allegedly working with elements within enforcement agencies (especially the Malaysian Anti-Corruption Commission [MACC], and possibly police) to pressure, intimidate or investigate corporate founders/executives — leading to forced exits, share sales or takeovers of companies.
One businessman at the centre of the controversy described it bluntly:
The “corporate mafia” is a method, not a particular group, used repeatedly across different companies.
There are also claims of:
• Payments framed as “fees” for protection;
For more stories about where money flows, click here for Capital Section
• Coordinated actions involving investigations, raids, or regulatory pressure; and
• Targeting of founder-led or family-controlled firms, where control can be wrested.
2. What has actually happened (facts, not claims)
Several concrete developments have taken place:
Investigations and raids
• Police, the Securities Commission Malaysia (SC) and other agencies raided premises linked to individuals in the probe; and
• Assets seized included cash, luxury items and documents.
Multi-agency probe
• Investigations involve police (AMLA [anti-money laundering] task force), SC, MACC, Inland Revenue Board
Political escalation
• Calls for a Royal Commission of Inquiry (RCI) into MACC conduct
Government response
• A special committee was formed to examine allegations against the MACC leadership
3. Strong denials from institutions
Critically, none of the allegations have been proven in court.
• MACC has categorically denied involvement; and
• The MACC chief has filed legal action against media reports.
The official position: Allegations are “baseless” and meant to discredit enforcement actions.
4. Why this story matters (beyond the headlines)
Even if unproven, the implications are enormous.
(A) Weaponisation of enforcement power
If true, this suggests:
• Enforcement without rule of law; and
• Enforcement is a strategic tool in corporate battles.
That is qualitatively different from normal corruption.
(B) A new tool of corporate control
Traditionally, takeovers rely on:
• Capital;
• Markets; and
• Shareholder vote
This allegation implies a different model:
Regulatory pressure → forced capitulation → asset transfer
That is closer to coercive capitalism, not market capitalism.
(C) Malaysia’s structural vulnerability
Malaysia has long had:
• Strong state-linked institutions;
• Concentrated ownership structures (family firms, government-linked companies); and
• High regulatory discretion.
This creates a risk where:
Control of institutions is control of outcomes.
(D) Investor confidence risk
For markets, the real issue is not whether this specific case is true.
It is whether investors believe:
• Corporate control can be arbitrarily influenced; and
• Enforcement is predictable and neutral.
If not:
a. Risk premium rises;
b. Founders avoid listing; and
c. Capital migrates elsewhere.
5. Historical context (this is not entirely new)
Malaysia has seen:
• Politically linked corporate battles;
• Regulatory interventions in corporate disputes; and
• High-profile corruption and enforcement cases.
But what is new here is the explicit allegation of a repeated “playbook”.
Enforcement → pressure → takeover
This is why the term “corporate mafia” resonates — it suggests systemisation, not isolated abuse.
6. The competing narratives
This issue is now split into two opposing interpretations:
Narrative 1: Abuse of power
• Enforcement agencies are being captured or influenced; and
• Used selectively in corporate struggles.
Narrative 2: Pushback against enforcement
• Powerful business interests are fighting back against investigations; and
• Allegations are part of a counter-narrative to weaken institutions.
Both cannot be true at the same time — but both can be partially true.
7. The underlying question
This is not really about one businessman or one agency.
It is about a deeper issue:
Can institutions in Malaysia act independently of economic and political power?
Because if they cannot:
• Capitalism becomes negotiated, not competitive; and
• Ownership becomes contested through influence, not price.
The bottom line
This is not just a scandal. It is a stress case.
• If the allegations are false, institutions must prove it transparently; and
• If the allegations are true, the system is far more fragile than assumed.
Either way: Markets cannot function where enforcement is seen as a weapon.
It is a long, comprehensive report on the facts, the actions and reported responses, and the implications, even if they have yet to be proven in court. The central argument is that the weaponisation of enforcement power is itself a form of corruption. This, however, is not entirely new. Malaysia has a history of politically linked corporate battles, regulatory interventions and high-profile corruption and enforcement cases — it is impressive that ChatGPT picked up on this.
ChatGPT correctly argued that this is not about one businessman or one agency, but about whether institutions in Malaysia can act independently of economic and political power. At its core, the question is whether Malaysia operates under coercive capitalism or market capitalism. In this respect, the country is especially vulnerable, given its extensive state-linked institutions in business, concentrated ownership structures and high regulatory discretion.
It is a highly impressive report — arguably better than much of what has been written elsewhere. But does it replace journalism altogether?
My concise summary of the three articles published in The Edge
1. The fight is about a RM2.5 billion government contract to supply Malaysian passports and MyKad — a highly sensitive and long-term national security business.
2. The core issue is politics and contracts: NexG (formerly Datasonic) has long depended on government contracts, where connections often matter as much as capability, if not more so. Because of the frequent changes in government since 2018, these contracts became more unstable, forcing constant realignment with new political power centres.
3. The power struggle: After securing major contracts in 2025, a boardroom battle erupted. The dispute centres on whether NexG’s core technology should be outsourced (allegedly to HeiTech Padu). Multiple factions emerged — NexG co-founder, executive chairman and CEO Datuk Abu Hanifah Noordin and businessman Victor Chin, and a rival group allegedly linked to political figures.
4. “Corporate mafia” allegation: claims of a network using intimidation, coercion and enforcement agencies (police AMLA task force, MACC) to gain control of the company.
5. The allegations include pressure to transfer shares, freezing of bank accounts, use of investigations as a leverage. There has been no conclusive evidence yet, with heavily contested narratives. The Edge chronicled the actions of the various enforcement agencies — the facts.
6. Money, influence and intermediaries: An alleged RM9.5 million payment was made to a politician to “resolve issues”. Shadowy figures (Mr A, Mr R, Mr X) were said to influence contracts, board decisions and share transfers.
7. The identities and background of all the players (including Mr A, Mr R and Mr X) are either named or implied — and how they are connected to each other.
8. Outcomes: Control appears to have shifted back to a Hanifah-aligned faction after board resignations. But investigations continue, accounts remain frozen and legal/regulatory risks persist.
9. The bottom line: this is not just a corporate dispute — it is how political access, regulatory power and corporate control intersect in Malaysia’s government-linked business ecosystem.
Do you see the difference between the reports by ChatGPT and The Edge?
Information versus judgement
If ChatGPT had been around between 2013 and 2016, when The Edge reported on the 1Malaysia Development Bhd (1MDB) scandal, it would likely have produced a similar analysis. It would have assembled the available facts: what 1MDB was, what it was accused of and the competing narratives.
On one side were The Edge’s reports alleging that funds had been misappropriated from 1MDB through entities and individuals including PetroSaudi, Tarek Obaid and Low Taek Jho (Jho Low) — allegations that had not yet been proven in court at the time. On the other side, those accused, together with the Malaysian government, denied the allegations, arguing that they were part of an attempt to overthrow the government using fake news and tampered evidence derived from stolen emails. The whistle-blower was also arrested and later jailed for extortion.
ChatGPT would never have recirculated the money trail chart (published in The Edge Financial Daily and reproduced here), with its precise details of amounts, dates, banks, account numbers and account holders — even though it was widely regarded as central evidence for enforcement agencies worldwide. It was, in effect, the smoking gun.
This limitation is because AI does not decide. It takes no position on right or wrong, and it does not “believe” anything.
If you are still unclear, let me now explain. This is what Part 8 is all about — the difference between information and judgement!
AI can assemble the world. Journalism decides what it means.
The rise of AI has collapsed the cost of knowledge. It can be comprehensive, structured, even persuasive — pulling together facts, counterarguments and historical context in seconds. It is, in many ways, the ultimate research assistant: tireless, multidimensional and increasingly indistinguishable in tone from human writing.
But it stops at the end of commitment.
AI does not believe. It does not risk being wrong. It does not carry consequences for the positions it presents. It synthesises — but it does not stand.
Journalism, at its best, does.
Good journalism is not merely the arrangement of facts; it is the exercise of judgement. It decides what matters, what is noise and what is deliberately obscured. It provides context not just to inform, but to clarify intent — who benefits, who loses and why something is happening now. It connects dots that are not always visible in data, but are understood through experience, sources and institutional memory.
And here is where AI changes the equation — not by replacing journalists, but by amplifying them.
AI is not the competitor but the multiplier
AI makes journalists work faster. It broadens their reach, deepens their research and sharpens their ability to evaluate arguments from multiple angles. Journalists need to not just ask the questions but question the answers. Sources are inherently biased. And repeatedly using the same source threatens the journalist’s independence. AI also removes the mechanical burden of gathering and organising information, freeing up time for what truly matters: thinking, judging, deciding.
In that sense, AI is not the competitor but the multiplier. And this creates a new divide — not between humans and machines, but between those who use AI and those who do not. A journalist armed with AI will out-research, out-write and outpace one who resists it. The gap will not be marginal; it will be decisive.
And yet, even at its most powerful, AI still cannot cross the finish line.
More importantly, journalism has a name behind it.
Turning brand into identity by building credibility and trust
A journalist signs a story. An editor stands over it. A publication stakes its credibility on it. That act — of attaching identity to interpretation — is what gives journalism its weight. It is accountable. It can be challenged, proven wrong or vindicated over time. That is what builds trust.
AI, by contrast, is infinitely adaptable — and, therefore, infinitely non-committal. It can generate multiple perspectives, but it does not choose one. It can outline the debate, but it does not resolve it. It informs the conversation, but it does not lead it.
AI cannot build credibility over time or earn trust through consistency of judgement. It cannot decide, in moments of ambiguity, which side of the argument it believes is right — and defend that position. Journalism can.
And in a world saturated with information, leadership of thought matters more than ever. More than intellectual perspectives. More than historical background.
The value of journalism, therefore, is not diminished by AI — it sharpens it. When facts are abundant, judgement becomes scarce. When everyone can access information, credibility becomes the differentiator. When narratives can be generated endlessly, the voice that is trusted to take a stand becomes the asset.
AI will dominate the supply of content. Journalism will define its meaning. The winners will be those who understand both.
The future will not belong to those who can say the most, but to those who are willing to say, clearly and credibly, what they believe is true.
Truth needs a voice. AI has none.
Portfolio commentary
The Malaysian portfolio gained 1.4% for the week ended July 8, with all stocks in positive territory save for Hong Leong Industries, which was unchanged. The biggest winners were United Plantations (+4.1%), Public Bank (+3.8%) and Malayan Banking (+2.6%). Total portfolio returns now stand at 225.3% since inception. This portfolio is outperforming the benchmark FBM KLCI, which is down 8% over the same period, by a long, long way.
The Absolute Returns Portfolio also performed well, up 1.2% for the week. The gains lifted total portfolio returns to 26.3% since inception. The biggest gainer was Alibaba Group Holding (+15.8%) while Nvidia and Talen Energy gained 3.3% and 2% respectively. Schneider Electric (-4.4%), Berkshire Hathaway (-1%) and Microsoft (-0.2%) were the three losing stocks last week.
The AI Portfolio, on the other hand, fell 2.4% on a continued sell-off in chip stocks, paring total portfolio returns to 26.4% since inception. The top losers were Marvell Technology (-14.8%), Naura Technology (-14.4%) and Unusual Machines (-11.2%). Meanwhile, the biggest gainers were Alibaba (+15.8%), Akamai Technologies (+12.2%) and Broadcom (+5.2%).
Disclaimer: This is a personal portfolio for information purposes only and does not constitute a recommendation or solicitation or expression of views to influence readers to buy/sell stocks, including the particular stocks mentioned herein. It does not take into account an individual investor’s particular financial situation, investment objectives, investment horizon, risk profile and/or risk preference. Our shareholders, directors and employees may have positions in or may be materially interested in any of the stocks. We may also have or have had dealings with or may provide or have provided content services to the companies mentioned in the reports.
