As I prepare to launch a new brand under the AVNetwork umbrella, Workplace AI (more on that soon), I am consuming more content on the topic of AI than ever before—if that's even possible.
On Oct. 21, in the midst of prepping an article to post on AVNetwork.com, our CMS went offline because of the worldwide PHP vulnerability. Cybersecurity firm Calif and cloud security firm Wiz discovered the vulnerability using AI tools from OpenAI and Google. We've come to know that it affected approximately 70 percent of websites globally. Miraculously, thanks to our engineering and IT teams, I was back up and running in less than an hour.
The world is still abuzz when, in September, ex-Anthropic employee Jacob Coxon declared on X, “The people building AI earnestly believe that it could kill us all by the end of the decade.” Then, to put a finer point on it, Evan Hubinger, a current team lead at Anthropic, responded, “Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.”
As I plan and create content for Workplace AI, I am focused on the benefits of the here and now, and I am excited.
Coincidentally, on Oct. 21, Anthropic's co-founder, president, and chair of the board of directors, Daniela Amodei, posted to LinkedIn the foreword she wrote to the company's recently released charter, Anthropic Public Benefit Report.
The report comprises five pillars, with the first pillar being, “The technical problems of AI safety are solved,” which is a particularly bold objective.
In her foreword, Amodei touted the many successes organizations have reported using Claude, such as: “In the lab, Claude designed new protein binders from scratch. In Burkina Faso, a team with the International Rescue Committee asked Claude to turn clinic registers into digital dashboards. Those dashboards flagged a spike of 25,000 pneumonia cases, so the team can anticipate and plan for future surges. A lawyer in India who feared the dense prose of Shakespeare began reading 'Hamlet.'"
Amodei is quick to point out the potential and real pitfalls: “AI is a dual-use technology. The same intelligence that powers protein binder discoveries could be used by malicious actors to engineer biological weapons. Without appropriate safeguards and release structures, cyber capabilities could be exploited to carry out large-scale attacks instead of serving to strengthen our infrastructure. If not properly trained and carefully deployed, AI can be a danger in its own right.
For these reasons, Anthropic has a duty to ensure our models are robustly safe and aligned with human values and goals.”
Personally, I am holding out hope that companies such as Anthropic, OpenAI, and others are able to build models that ensure humanity first.
The Safety Pillar
The full 20-page report is worth reading, but the transparency outlined in detail by Anthropic in the first pillar is sobering and should be food for thought for all of us.
Pillar 1: The Technical Problems of AI Safety are Solved, ends with the following statement: “The same safeguards that prevent our models from being exploited by attackers and other malicious actors can sometimes also prevent the beneficial work of defenders and researchers. Our bio and cyber verification programs are meant to reduce these roadblocks for verified users, but the review process can take days or weeks, and the programs are not yet available on every plan or platform. We have made improvements to our safeguards that have reduced false positives, but many legitimate researchers still encounter blocks.”
Read the Anthropic 2026 Public Benefit Report.
Today, Tomorrow, and
If there’s such a thing, I am a cautious Pollyanna. I am hopeful that humanity will win.
And in my current world, writing about, illuminating stories, and helping to enlighten my readers about how AI can help enhance the collaborative workplace (which is anywhere you work) is what has me super excited. Stay tuned!
