Michael Akinwumi
banner
datawumi.bsky.social
Michael Akinwumi
@datawumi.bsky.social
Chief AI Officer, Head of Responsible AI Lab & Rita Allen Alum. @datawumi on X. Advisor. Personal Opinion.
On #VeteransDay, we honor those who remind us that people are the true infrastructure of progress.

As veterans power our war and defense systems, people must also power our AI infrastructure — because without human values, even the smartest code fails.
November 11, 2025 at 9:08 PM
I admire Watson’s science. I truly do.
But if he could map DNA and still miss that racism isn’t genetic, then the flaw wasn’t in the code — it was in the interpretation.

Let’s celebrate discovery, not prejudice dressed up as data. #Science #CivicScience
November 8, 2025 at 7:03 AM
November 8, 2025 at 7:03 AM
But trust without transparency or accountability is risky especially now that 82% of Americans use AI for #housing insights as shown in @natfairhouse.bsky.social's 2025 Trends Report.
November 7, 2025 at 12:04 PM
Without oversight, AI can recreate the same discriminatory patterns we saw in redlining now that the technology is nearly as trusted as real estate agents.

That’s why #AI is now a civil rights and existential rights frontier—not just a #tech issue.

#AIethics #HousingJustice
November 7, 2025 at 12:04 PM
AI is now nearly as influential as real estate professionals. It’s shaping how people find homes, qualify for loans, and understand property values.
But #tech can scale harm just as easily as it scales access.
@natfairhouse.bsky.social has seen this before. #AI #FairHousing
November 7, 2025 at 12:04 PM
@natfairhouse.bsky.social's s 2025 Trends Report finds #housing discrimination complaints based on national origin are at their highest level since 2018.
November 6, 2025 at 9:40 AM
November 5, 2025 at 12:46 AM
This experience strengthens my belief that localized small language models (SLMs) — not massive LLMs or AGIs — deserve more resources, investment, and research.

The country that masters local intelligence will ultimately win the AI race.

Agreed?
#SLMs #LanguageTech
October 25, 2025 at 3:20 AM
Tried using Meta AI to privately process a short Yoruba message before texting my mom — it just kept repeating the same line.

This further shows that we still lack language models that understand local languages, context, and culture.
October 25, 2025 at 3:20 AM
AI infrastructure is booming in India. As India & the U.S. deepen tech ties, key questions emerge:
(1) How do we build trusted AI ecosystems?
(2) How do cross-border partnerships drive innovation within global laws?
Join me for the conversation
Registration: docs.google.com/forms/d/e/1F...
October 23, 2025 at 2:59 PM
“Blessed are the peacemakers, for they shall be called children of God.”
May the uncertainties that lie ahead give way to lasting peace. And may that same spirit of peace extend to our own institutions—where stability begins with turning on the regulatory lights in our nation.
October 13, 2025 at 1:11 PM
If you’re a human model thinking your job’s safe from #AI — think again.

This is “me” modeling in Vegas, promoting AI Agents. The twist? I never signed a contract — Sora did the work.

(Would you know it wasn’t real?)

#FutureOfWork #DeepfakeEra #Misinformation
October 12, 2025 at 8:47 PM
October 10, 2025 at 1:34 PM
3/ Algorithms carry values—even silence is a choice. If we don’t assert plurality in AI design, we risk turning data into another resource to be mined.

This isn’t about choosing tech over culture. It’s about pushing for systems that let local voices set the lens.
October 4, 2025 at 3:41 AM
2/ If we define “truthfulness” from the Global North, aren’t we just exporting a worldview under the guise of accuracy?

What governance must the Global South build so that language, culture, and cognitive labor aren’t quietly extracted or colonized by AI systems?
October 4, 2025 at 3:41 AM
September 28, 2025 at 8:08 AM
I rediscovered my response to this Quorum question seven years later: “How can I help humanity achieve the greatest things in a cosmic scale, in my short life?” Believe you me, it further renews my vision. I keep changing.
September 28, 2025 at 5:35 AM
The ultimate question for #security experts, including #AI security researchers, is will the U.S.-TikTok deal solve the “national security” problems it was designed to overcome?
September 26, 2025 at 4:32 PM
Albania appointed Diella, an AI Cabinet Minister, to oversee public procurement—an experiment aimed at boosting transparency, accountability and curbing corruption.
#GovAI vs #AIGov via @economist.com
September 11, 2025 at 8:32 PM
Rain falls on every roof, without bias. Humans should do better.

When open debate dies, democracy withers.

In #AI too, dismissing fairness as “woke” shuts doors to innovation + equity.
Dialogue > division. Evidence > ideology.
#AIFairness #AI
September 11, 2025 at 7:06 PM
Do you want to learn more about #DataCenter conversation in #PGCounty MD? Read the Notice of Meeting of the Qualified Data Center Task Force in the County.
September 10, 2025 at 1:38 AM
A hint of what an economy powered by #AGI or #superintelligence might look like. An opportunity for humanity to reflect on what it truly means to "win the #AI race."
September 9, 2025 at 2:14 PM
If Comte is right, then to understand #AI we must know both its #math and its #sociology. Ignore the equations, and you miss the engine. Ignore #society, and you miss the steering. Claim “beneficial AI” without both, and your argument is stillborn.
September 5, 2025 at 5:51 PM
Forget the software race—physical AI is where supremacy will be decided. The country that cracks embodied intelligence first will define AGI and superintelligence for the rest of us. via @economist.com
August 16, 2025 at 1:16 PM