AI is not a new construct

𝐈'𝐕𝐄 𝐃𝐈𝐒𝐂𝐎𝐕𝐄𝐑𝐄𝐃 𝐖𝐇𝐀𝐓 𝐀𝐈 𝐆𝐎𝐕𝐄𝐑𝐍𝐀𝐍𝐂𝐄 𝐑𝐄𝐀𝐋𝐋𝐘 𝐈𝐒.

Codswallop. Absolute codswallop. - NO YOU HAVE NOT

𝐀𝐈 𝐡𝐚𝐬𝐧'𝐭 𝐞𝐱𝐩𝐨𝐬𝐞𝐝 𝐚 𝐠𝐨𝐯𝐞𝐫𝐧𝐚𝐧𝐜𝐞 𝐠𝐚𝐩. 𝐈𝐭'𝐬 𝐞𝐱𝐩𝐨𝐬𝐞𝐝 𝐚 𝐤𝐧𝐨𝐰𝐥𝐞𝐝𝐠𝐞 𝐠𝐚𝐩. 𝐈'𝐯𝐞 𝐛𝐞𝐞𝐧 𝐠𝐨𝐛𝐬𝐦𝐚𝐜𝐤𝐞𝐝 𝐛𝐲 𝐡𝐨𝐰 𝐦𝐚𝐧𝐲 𝐩𝐞𝐨𝐩𝐥𝐞 𝐢𝐧 𝐜𝐲𝐛𝐞𝐫, 𝐫𝐞𝐬𝐢𝐥𝐢𝐞𝐧𝐜𝐞, 𝐫𝐢𝐬𝐤 𝐚𝐧𝐝 𝐜𝐨𝐦𝐩𝐥𝐢𝐚𝐧𝐜𝐞 𝐬𝐭𝐢𝐥𝐥 𝐝𝐨𝐧'𝐭 𝐮𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝 𝐰𝐡𝐚𝐭 𝐠𝐨𝐯𝐞𝐫𝐧𝐚𝐧𝐜𝐞 𝐢𝐬.

Every day I see another six posts claiming AI has created a brand-new form of governance - IT HAS NOT and it never will period.

Get over it your new AI governance agent (that you believe is changing the world) is doing nothing at all.

Apparently...

1️⃣ "𝐀𝐈 𝐧𝐞𝐞𝐝𝐬 𝐚 𝐧𝐞𝐰 𝐉𝐮𝐝𝐠𝐞𝐦𝐞𝐧𝐭 𝐋𝐚𝐲𝐞𝐫."

No. AI may influence judgement. Governance determines accountability.

2️⃣ "𝐀𝐈 𝐰𝐢𝐭𝐡𝐨𝐮𝐭 𝐠𝐨𝐯𝐞𝐫𝐧𝐚𝐧𝐜𝐞 𝐢𝐬 𝐝𝐚𝐧𝐠𝐞𝐫𝐨𝐮𝐬."

Codswallop. AI isn't "without governance". Organisations either govern it properly or they don't. The principles haven't changed.

3️⃣ "𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐀𝐈 𝐫𝐞𝐪𝐮𝐢𝐫𝐞𝐬 𝐀𝐠𝐞𝐧𝐭 𝐆𝐨𝐯𝐞𝐫𝐧𝐚𝐧𝐜𝐞."

No. Sandboxes, policy engines, safety evaluators and verifiers are engineering controls. They are governed. They are not governance.

4️⃣ "𝐀𝐈 𝐌𝐞𝐝𝐢𝐚𝐭𝐢𝐨𝐧 𝐒𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐢𝐬 𝐭𝐡𝐞 𝐧𝐞𝐱𝐭 𝐞𝐯𝐨𝐥𝐮𝐭𝐢𝐨𝐧 𝐨𝐟 𝐠𝐨𝐯𝐞𝐫𝐧𝐚𝐧𝐜𝐞."

No. That's architecture, assurance and control engineering with a new marketing label.

5️⃣ "𝐀𝐈 𝐆𝐨𝐯𝐞𝐫𝐧𝐚𝐧𝐜𝐞 𝐢𝐬 𝐚 𝐛𝐫𝐚𝐧𝐝-𝐧𝐞𝐰 𝐝𝐢𝐬𝐜𝐢𝐩𝐥𝐢𝐧𝐞."

No. AI is simply another domain to which governance is applied. Accountability, decision rights, oversight, assurance and evidence existed long before AI.

This is becoming a cottage industry. Every week another framework, maturity model and consulting proposition appears, all trying to convince us AI has reinvented governance.

It hasn't.

Governance still asks exactly the same questions it always has.

Who is accountable?

Who owns the decision?

Who accepts the risk?

Who provides independent assurance?

What evidence supports the decision?

Technology evolves.

Governance extends into AI.

𝐈𝐭 𝐝𝐨𝐞𝐬 𝐧𝐨𝐭 𝐫𝐞𝐢𝐧𝐯𝐞𝐧𝐭 𝐢𝐭𝐬𝐞𝐥𝐟 𝐛𝐞𝐜𝐚𝐮𝐬𝐞 𝐬𝐨𝐦𝐞𝐨𝐧𝐞 𝐜𝐨𝐢𝐧𝐞𝐝 𝐚𝐧𝐨𝐭𝐡𝐞𝐫 𝐛𝐮𝐳𝐳𝐰𝐨𝐫𝐝 𝐨𝐧 𝐋𝐢𝐧𝐤𝐞𝐝𝐈𝐧 or elsewhere.

#AIGovernance #ArtificialIntelligence #CorporateGovernance #RiskManagement #Leadership