cigitalgem

cigitalgem, 1 day ago to ML

When you choose to use an #ML #LLM foundation model, you accept the risk management decisions made by the vendor without your input. Wonder what they are? Read this #MLsec paper from #IEEE computer.

https://berryvilleiml.com/2024/05/16/how-to-regulate-llms/

cigitalgem, 3 days ago to random

Oh look. Data feudalism is everywhere. #MLsec https://www.engadget.com/yuck-slack-has-been-scanning-your-messages-to-train-its-ai-models-181918245.html?src=rss&utm_source=press.coop

cigitalgem, 4 days ago to random

BIML #MLsec

cigitalgem, 4 days ago to ML

I am speaking tonight at the #ISSA NOVA chapter meeting. Meeting starts at 5:30 in Reston at the Microsoft building.

10, 23, 81 — Stacking up the LLM Risks: Applied Machine Learning Security

https://www.issa-nova.org/may-16-530pm-dr-gary-mcgraw-on-stacking-up-the-llm-risks-applied-machine-learning-security/

#MLsec #ML #AI #LLM

cigitalgem, 4 days ago to ML

Wonder how the goverment should regulate LLMs? Here's how. #MLsec #ML #AI #LLM

https://berryvilleiml.com/2024/05/16/how-to-regulate-llms/

cigitalgem, 5 days ago to llm

Giving my #LLM Risks talk (BIML work) today for Google on a global forum. If you work for Google, hop on! #MLsec #ML #AI

Noon Eastern....hosted by Zurich.

cigitalgem, 5 days ago to random

Data feudalism is alive and well. Land grab continues with all of its legal entanglements. https://www.nytimes.com/2024/04/30/business/media/newspapers-sued-microsoft-openai.html

cigitalgem, 8 days ago to llm

Recursive pollution is a very serious #LLM #MLsec risk. At BIML, we slate it as number one. DO NOT EAT YOUR OWN BRAINS.

#ML #AI

https://www.ft.com/content/053ee253-820e-453a-a1d5-0f24985258de

cigitalgem, 10 days ago to ai

CalypsoAI produced a video for RSA abojut #MLsec #AI #ML #security #infosec #LLM

Watch as I interview Jim Routh and Neil Serebryany.

https://berryvilleiml.com/2024/05/10/video-interview-a-deep-dive-into-generative-ai-and-cybersecurity/

cigitalgem, 12 days ago to random

Nothing to see here, just armed robots https://arstechnica.com/gadgets/2024/05/robot-dogs-armed-with-ai-targeting-rifles-undergo-us-marines-special-ops-evaluation/

cigitalgem, 14 days ago to infosec

More on the Synopsys SIG spin off #swsec @appsec #infosec
https://www.msn.com/en-us/money/savingandinvesting/synopsys-sells-software-integrity-unit-for-2-1-billion-to-pe-group/ar-BB1lUNfH

cigitalgem, 24 days ago to random

Proud to serve as an Advisor to Calypso AI, #MLsec made real.

It's not just a job, it's a wardrobe.

cigitalgem, 1 month ago to random

We just lost another great light of rationalism. Dan Dennett helped get me started in philosophy of mind way back in the late '80s. Dan was right about lots of things. https://dailynous.com/2024/04/19/daniel-dennett-death-1942-2024/

cigitalgem, 1 month ago (edited 1 month ago) to llm

I am giving a Machine Learning Security talk in Bergen, Norway Monday 22.4. The scandanavian tour's last and final stop. #MLsec

I will be discussing #LLM security risks and the new BIML results. Please join us. #ML #LLM #AI

Thanks for your help spreading the word.

https://www.lyyti.fi/reg/CDR-NO-22-04-2024

cigitalgem, 1 month ago to infosec

Thanks Stockholm. The breakfast seminar on #swsec was good. Next up is OSLO tomorrow morning (THURSDAY). If you are in Norway, please come join me!

I will also briefly cover machine learning security #MLsec

#appsec #infosec #software #LLM

https://www.lyyti.fi/reg/CDR-NO-18-04-2024

cigitalgem, 1 month ago to infosec

Software Security Seminar in Stockholm TOMORROW 17.4

Please join me for an early morning breakfast seminar on #swsec (with some #MLsec thrown in for good measure). Build security in.

Register here https://www.lyyti.fi/reg/CDR-SV-17-04-2024

#infosec #appsec #ML #LLM

Thank you in advance for passing this on to dev types you know in Sweden. Please boost for reach.

cigitalgem, 1 month ago to llm

NEW Machine Learning Security talk in Bergen, Norway Monday 22.4. The scandanavian tour continues!

I will be discussing #LLM security risks and the new BIML results. Please join me. #ML #LLM #AI

https://www.lyyti.fi/reg/CDR-NO-22-04-2024

cigitalgem, 1 month ago to random

Totality in Indiana on a boat (pictures, videos, and commentary).

https://apothecaryshed.com/2024/04/12/total-eclipse/

cigitalgem, 1 month ago to infosec

I am giving two #swsec breakfast seminars back to back mid-April. If you are in Sweden, Norway or Finland, please consider coming. Pass it on to those who may be interested.

#appsec #infosec #MLsec

STOCKHOLM 17.4 https://www.lyyti.fi/reg/CDR-SV-17-04-2024

OSLO 18.4 https://www.lyyti.fi/reg/CDR-NO-18-04-2024

cigitalgem, 1 month ago to llm

I am giving a talk @indianauniv in Bloomington THIS FRIDAY. I will cover #LLM security risks (that is #MLsec) identified by BIML. Open to the public.

https://spice.luddy.indiana.edu/garymcgrawtalk/

picture

cigitalgem, 1 month ago to random

Ross Anderson's first Silver Bullet episode (number 13 from 2007) was the most popular episode I ever recorded (out of 153 monthly episodes in a row). Ross recorded a second (number 70 in 2012). Here they are, moved to archive in 2018.

https://apothecaryshed.files.wordpress.com/2018/10/silverbullet-070-ross-anderson-2.mp3

https://apothecaryshed.files.wordpress.com/2018/10/silverbullet-013-ross-anderson.mp3

cigitalgem, 1 month ago to random

Ross Anderson @rossjanderson was one of the best security engineers on the planet and a friend. He will be sorely missed.

https://en.m.wikipedia.org/wiki/Ross_J._Anderson

May you go forth with the splendor of fire, with the speed of lightning, with the swiftness of wind;

cigitalgem, 1 month ago to ai

For ML, open source must include training data, or it's not a useful distinction. The machine becomes the data.
#datafeudalism #MLsec #AI #LLM

https://www.technologyreview.com/2024/03/25/1090111/tech-industry-open-source-ai-definition-problem/

cigitalgem, 1 month ago to ML

Nice to see data lakes released...but what we need are data oceans. This new dataset is off by many orders of magnitude. Humans have a hard time with trillions...
#MLsec #ML #AI #LLM #datafeudalism https://huggingface.co/blog/Pclanglais/common-corpus

cigitalgem, 1 month ago to ML

I don't believe we can filter our way out of drinking a polluted ocean of training data. #MLsec #ML #AI #LLM https://www.techtarget.com/searchEnterpriseAI/news/366574580/Microsoft-hires-DeepMind-co-founder-amid-Google-Apple-news