Over the past six months, AI labs including OpenAI, Anthropic and Abu Dhabi‑based G42 have approached firms in Dubai, Riyadh and Abu Dhabi with offers ranging from $3 million to $7 million for access to Slack archives, internal email logs and Teams chats. The deals are framed as training material for large language models, with labs promising anonymized aggregation but retaining raw metadata.

In response, several UAE free‑zone companies and Saudi Aramco‑linked contractors have signed non‑disclosure agreements to review the proposals, while others have declined citing the UAE’s Personal Data Protection Law (PDPL) and Saudi Arabia’s PDPL, which require explicit employee consent for processing personal communications. No public disclosure of the exact number of deals has been made, but industry sources estimate at least twelve mid‑size enterprises have received formal offers.

The trend exposes a gap between rapid AI development and regional data‑governance frameworks that are still maturing. If labs succeed in harvesting Slack DMs at scale, they could unlock nuanced workplace behavior patterns that improve model accuracy but also increase the risk of credential leakage, insider threat profiling and unauthorized surveillance.

From a business standpoint, firms that accept the payments may gain short‑term liquidity but risk reputational damage and potential regulatory fines under PDPL, which can reach up to 4 % of global turnover. Conversely, those that refuse may lose out on a nascent revenue stream as AI labs look elsewhere for rich, domain‑specific language data.

AI labs such as OpenAI, Anthropic and G42 stand to win by acquiring unique, high‑frequency corporate dialogue that improves model grounding in Gulf‑specific business contexts. Large multinationals with robust legal teams may win by negotiating favorable terms or leveraging the offers as a bargaining chip for better AI service pricing.

Mid‑size tech firms and startups in Dubai Internet City or Riyadh’s Tech Valley are the most likely losers, lacking the resources to negotiate strong data‑use clauses and facing pressure to accept cash offers that could breach employee trust. Regulators in the UAE and Saudi Arabia could also lose if enforcement lags, undermining confidence in their nascent data‑protection regimes.

Watch for the first PDPL enforcement actions related to AI‑data purchases, likely to emerge from the UAE Data Office or Saudi Data & AI Authority within the next 12 months. Also monitor whether AI labs begin offering opt‑out mechanisms or financial incentives directly to employees, a shift that could redefine the consent model. Finally, track any movement toward regional data‑trusts or industry‑wide consortia that negotiate bulk data licenses on behalf of member firms.