On 11 November 2026, Northern Cape police spokesperson Capt. Nomthandazo Mnisi warned residents about festive-season investment scams, bank impersonation, online-dating fraud and fake-prize schemes, and urged victims to report to police. It is a routine warning. It is also an invitation to report crimes to an institution whose capacity to investigate them is now in question.
The strongest case for the current framework
The Cybercrimes Act 19 of 2020 was a necessary reform. It replaced a patchwork of common-law fraud and outdated electronic-communications offences with defined crimes, including cyber fraud, forgery and unlawful access. Concentrating investigative responsibility in one national police service is also defensible. Specialist skills are scarce, evidence-handling standards need to be uniform, and victims need a single place to report. Section 54 adds a duty on electronic communications service providers and financial institutions. Where they are aware their service or network is involved in an offence in a prescribed category, they must report it to the South African Police Service "without undue delay and, where feasible, not later than 72 hours" and preserve information that may help the investigation (Cybercrimes Act, s 54). A provider that fails to comply faces a fine of up to R50 000. That is the design of a law that expects police to act on what they receive.
What the capacity numbers show
The expectation runs into a staffing problem. In a June 2026 statement, the Democratic Alliance said a parliamentary reply showed that "not a single detective stationed at the Top 35 highest-crime police stations received digital forensics training" between 1 April 2025 and 27 February 2026. Only 15 detectives at those stations received financial-investigation training in the same period (DA statement via allAfrica). The same statement says those stations have 3 496 funded detective posts against an approved requirement of 4 607, and only 2 480 of the funded posts are filled.
These figures reach us through an opposition press release. I could not read the reply document itself, so they should be treated as the DA's account of it. The release's claim that the gap is a "deliberate decision" is the party's characterisation and is not shown by the figures. The Top 35 grouping is real. SAPS uses a "30 plus 5" principle for its highest-contact-crime stations, and the Acting Minister confirmed there are 35 of them nationwide, 15 in Gauteng (parliamentary reply, Question NO512).
SAPS does have training plans. A separate reply to Parliament sets out the 2026/27 plan: 764 detective and crime-investigation courses for 13 576 learners, against 44 cybercrime courses for 770 learners (parliamentary reply, Question NW2577). That reply says the Directorate for Priority Crime Investigation supports specialised training in cybercrime investigations and digital forensics, but it gives no numbers. This is a plan for the coming year. It does not show how many detectives at the busiest stations are trained today. The cybercrime courses are roughly 6% of the courses planned for detective and crime-investigation training, and 770 learners is a very small base for a country where fraud now arrives mostly through phones.
Why this matters for the law
The policy lesson is not that South Africa needs a harsher statute. A scam victim in Kimberley or Upington usually goes first to a local station. If the detective there cannot preserve a seized phone, request data from a platform or bank, or document the chain of custody, the victim gets a case number and little else. The Act's offences then exist on paper while enforcement is thin where it matters most.
There is a second gap. Section 54 reporting applies only to offence categories that the responsible Cabinet member prescribes by Gazette notice, together with the form and manner of reporting (s 54(2)). The statute text I read shows no such notice, and I found none in my searches, though I could not confirm the section's commencement status. If the notice has not been made, the 72-hour duty on banks and telecoms has nothing to attach to. Putting duties on private firms before police can act on the reports they would generate only adds compliance cost.
A proportionate way forward
A pro-innovation reading of the problem avoids two bad responses. One is to widen police and platform powers, through broader monitoring obligations or intrusive access to data, because enforcement is weak. Section 54(4) rightly says providers need not monitor data or actively seek out unlawful activity, and that safeguard should stay. The other response is to treat the Act as working because it is on the books.
The better path is boring and measurable:
- Train first, at the stations that take the reports. Publish annual figures on digital forensics training for detectives at the highest-volume stations.
- Make section 54 operational or say why it is not. Publish the prescribed categories and reporting form, or give a timetable, so banks and carriers know what they must report and police know what to expect.
- Use partnerships that already exist. Banks and telecoms have fraud-analytics teams. A structured referral channel, with clear evidence standards, would help overstretched detectives more than new offences would.
- Report outcomes, not warnings. Awareness campaigns like the Northern Cape one are cheap and useful. Their value rises when the public can also see how many reported scams reached an investigator with digital skills.
The Cybercrimes Act sets the right goal. A festive-season warning to "report to police" is only as credible as the detective who picks up the report.