Exclusive: We obtained secret tapes of an emergency NRA meeting held shortly after the Columbine shooting that reveal the group’s PR strategy.

Officials considered a victims fund — but worried it would be “crass” and chose an unapologetic stance

The conference call was convened so top NRA officials could decide whether to cancel their 1999 Denver convention, scheduled just a few miles away from the site of the mass shooting that left 13 dead and 20 injured.

Some agonized over the optics.
One PR adviser worried that canceling the convention would result in the NRA’s most extreme members descending on the Denver area — and top officials derided those members as “hillbillies” and “fruitcakes” who might go off-script after Columbine and embarrass them.
At one point in the call, an NRA official wondered if the group should set up a $1 million fund for Columbine victims — a more sympathetic posture. But other participants worried that would make it seem like they were accepting responsibility:
The secret tapes show NRA officials wanted to avoid a firestorm like the one that ensued after the 1995 Oklahoma City bombings, which led to former President George H.W. Bush publicly resigning from the group — and resulted in the exodus of half a million members.
Officials on the call also dismissed politicians and industry heads as largely inconsequential.
CEO Wayne LaPierre said a GOP senator had asked for secret “talking points.” And a lobbyist reassured the others that the firearm industry would follow their lead:
The NRA officials ultimately chose not to cancel the convention because it would give the impression that the group had been “brought to its knees.”

That unyielding tone would become its standard messaging after mass shootings.
Read more revelations unearthed by the secret tapes here: https://t.co/ZdydNIUylk

More from All

How can we use language supervision to learn better visual representations for robotics?

Introducing Voltron: Language-Driven Representation Learning for Robotics!

Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z

🧵👇(1 / 12)


Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.

Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)

The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (
https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).

The secret is *balance* (3/12)

Starting with a masked autoencoder over frames from these video clips, make a choice:

1) Condition on language and improve our ability to reconstruct the scene.

2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)

By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.

Why is the ability to shape this balance important? (5/12)

You May Also Like