Senate Environment and Communications Legislation Committee, , 2:13:08 to 2:21:03
The department addresses concerns regarding broken links and typographical errors in the age assurance technology trial report, denying any intent to mislead the committee while acknowledging technical flaws.
Topics: Artificial intelligence, Online safety
Speakers: Senator Fatima Payman, Mr Jim Betts, Ms Sarah Vandenbroek, Unknown Speaker
Thank you, Chair. Good morning. My line of questions are around the evidence that the Department gave in August around this the reference links in the report of the age assurance technology trial. Which is thank you. Thank you. But became broken later on. Now, having considered the archives that I presented to the department in questions on notice,
No, because the department was operating in good faith on the understanding that those citations were correct, which was the responsibility of the author of the report. The moment that it became apparent that there were typographical errors and flaws in terms of the citations, they were identified as erroneous and a corrigendum correcting all identified errors was published alongside the report on the department's website. There was no intention to mislead the minor technical problems with that with those references, which in no way altered the substance of that report or not known to us at the time that evidence was given. Completely correct.
Mackay provided some evidence, but in question IQ 260102, I presented the department with archived versions of web pages which show that an article cited as Weber at all contained a digital object identifier which pointed to the wrong article in 2024 as it does now. So does the department believe that this pointed to the right article in 2025 and has changed twice in two years?
As a Secretary of the Department, I tend not to proofread references in technical reports. If the question is, did the department mislead the Senate? Then that would require us to have knowingly presented incorrect information. Yes, there were errors in those in the report in terms of the six citations and a few typographical mistakes. That is not in dispute from us, and my officials operated with appropriate due diligence in correcting those errors as soon as they came to light. They may wish to add more to that.
Okay, so are we saying that so do you agree that it's changed twice in a year?
Like is there evidence to show that no, I'm saying that they were that they weren't incorrect when the report was handed to government?
If um in giving evidence, and I'm speaking generally here, um with the best of intentions, an official um is unaware of a matter which is relevant to an answer they've given the best that they can subsequently do is to act with speed to correct that on the public record. This happens periodically in Senate estimates and elsewhere. Um in this case, we've clearly acted to correct um any errors which were on the record, and therefore I don't think this could be characterised as anybody attempting to mislead the parliament.
I guess mistakes can happen.
So it's admitting that a mistake was taken place and which is what a corrigendum does,
the committee, based on the lack of information that the officials had that they weren't able to provide to the committee, that it could have accidentally misled the committee. Um again, it could be no fault of the officials, but again, it's the role and duty of officials to have that information at hand before coming to estimates. Um at least that's the expectation that we senators have. Um Vandenbroek, um you told the committee in relation to this. Uh it literally took one of my staff members half an hour to uh pull these archives together and comprehensively debunk what you told the committee. Um again, like how were you unaware that a CCS's claim could be easily disproven? We expect officials to be experts in getting this information to us.
Uh we asked ACCS about the links. We operated on good faith on what we were told by them. I wasn't aware that there was a way that we could go and check back to the situation of a year ago. I was made aware of that, and then the team went and did that work.
Why has the department taken so long to take accountability for this misleading evidence?
I understand we uh Mr. Betts wrote to the committee not long after the hearing, once we had confirmed we'd had staff go through and check all of the hundreds of links in the report so that we knew there weren't any others other than the ones that were brought to our attention. As has been said, some other minor typographical problems were found during that process, they were also. So corrected, but we wanted to do that in a very thorough way before we corrected the record so that we could do it once and do it comprehensively.
So to the committee, would you do you have anything to say in particular of like I'm basically what I guess I'll go back to what Mr. Mackay said and hopefully that's going to resolve a lot of this? Um when asked in August, if these errors represent lazy report writing, Mr. Mackay told the committee I don't agree that it represents lazy report writing. If they had been broken at the time the report was finalized and provided to the government, then I think I would agree with you that these are links that are now broken but were not at the time that the report was provided. Mr. Mackay. Mackay, yes. Given it has now been proven the links were broken when the report was handed over, is it now your view that this is lazy report writing?