Request for Feedback: 5-Year Reflections on Kenya’s Data Protection Act.

On 03/05/2024 10.03, Mildred Achoch wrote:
> Dear Benson,
>
> Thank you for the information. Perhaps hybrid solutions can be explored.
> Language data held by the government could be considered a public good.
> Language data produced by private sector, for example local radios, TV
> stations and online content creators, could bring additional revenue for
> the owners of the data. If compensation is embedded in the new models,
> then this could even encourage more creation of local language content.
> There is a great need for inter-ministerial collaboration.
>

Very few African countries have language policies. It is a sensitive
topic, but of importance as it influences school curricular and
government service delivery. The language users need to be actively
involved in the process and adapt the technologies to their needs. A
lot can probably be done without much compensation if people feel they
will derive value. Perhaps the following quote from the economics of
language [1] maybe of interest:
“majority of Africans are governed in a
language that they do not understand”

1) Weber and Ginsburgh, “The Economics of Language”
repec.cepr.org/repec/cpr/ceprdp/DP13002.pdf

> Regards,
> Mildred Achoch.
>
> On Friday, May 3, 2024, Benson Muite via KICTANet
> <kictanet@lists.kictanet.or.ke <mailto:kictanet@lists.kictanet.or.ke>>
> wrote:
>
> On 02/05/2024 13.24, Mildred Achoch via KICTANet wrote:
> > Thank you for this opportunity to contribute. Regarding the AI aspect,
> > there should be some policies regarding the use of data that will
> train
> > AI. For example, in the case of local languages, which data will be
> > used? If data by local language content creators is used, will they be
> > compensated?
>
> Many African languages are poorly resourced, some are also endangered.
> Language data should be considered a public good as it can enable use of
> AI in many settings.  It does also enable easier surveillance of people
> that use that language.  There does not seem to be a government policy
> for local languages – even materials for school curricular such as
> standard texts and dictionaries are few.  Use of local languages at
> county level is probably the easiest place to start as there would  be a
> sufficient concentration of active language users.  Languages do divide,
> but they also enable diversity in culture and approaches to problem
> solving.  While the ODPC has a role to play, probably the main
> governmental actor is the Ministry of Sports, Culture and National
> Heritage.  The greatest widespread use of local languages is on radio.
>
> >
> > Regards,
> > Mildred.
> >