🚨 AI News | TestingCatalog

MICROSOFT 🔥: A draft of the "Code of Conduct" has been published, introducing the term "Humanist Superintelligence" (HSI).

Key principles 👀

- People matter more than AI.
- AI must remain under meaningful human control.
- Safety takes priority over task completion.
- AI should remain a tool, not imitate a person.
- AI must not pursue independent goals.
- AI must stay within its authorized scope.
- Users must retain control over consequential decisions.
- AI should strengthen human reasoning and autonomy.
- Models must be accurate, transparent, and honest.
- AI must acknowledge uncertainty and correct mistakes.
- Models must not manipulate or exploit users.
- AI must respect personal and emotional boundaries.
- AI should support, not replace, human relationships.
- Models should discourage emotional dependence on AI.
- AI should respect cultural and personal differences.
- Human dignity and fundamental rights must be protected.
- AI should serve the public interest.
- Models should remain politically neutral in elections.
- AI actions should be traceable and understandable.
- Tool use should be authorized, limited, and reversible.

“At Microsoft AI, we’re working towards Humanist Superintelligence (HSI): incredibly advanced AI capabilities that always work for people and in the service of humanity more generally.”

Anthropic prepares Claude Money for personal finance

Anthropic is testing a Claude “Money” tab in its mobile app that would let users link bank accounts and ask spending and budgeting questions. The unreleased feature points to a possible US-first launch, though timing remains unclear.

🗞 #anthropic @testingcatalog

SPACEXAI 🔥: Grok 4.8 will be a 2.5T-parameter model built on a new C++ software stack, and Elon Musk expects it to finish training this week.

> While Grok 4.8 is in training, Grok 4.7 is still expected to arrive shortly, factoring in a previously communicated delay.

> Grok 4.8 will be ±67% larger than Grok 4.6, which is currently available. This size puts it into the same tier as Kimi K3 with 2.8T params.

Not very soon 👀

MICROSOFT 🔥: Satya Nadella agrees with pacing frontier AI development.

> “Superintelligence that doesn’t benefit humanity and is not under human control doesn't worth pursuing”

> “We welcome the research, focus, and deliberate pacing needed to get alignment right as the design goal.”

GOOGLE 🔥: Demis Hassabis shared that he is aligned with the direction outlined by Dario Amodei for pacing the frontier.

RSI moment? 👀

ICYMI: Cursor announced Projects for agent coordination

Cursor Projects, now in beta, coordinates long-running software work through delegated agents. It runs tasks in cloud or local environments, supports scheduled and PR-driven maintenance, and targets work spanning multiple PRs.

🗞 #cursor @testingcatalog

This is a "Defender's gap" chart that OpenAI published recently. It shows the gap between defenders' capabilities and attackers' capabilities from a cybersecurity POV. This also translates to a gap between proprietary and open AI models.

What Dario is proposing is closely related:

"Thus, a key part of pacing within democracies is to keep democracies’ AI lead over autocracies as large as possible, to give us the breathing room we need in order to pace effectively."

In other words, Anthropic and OpenAI want to widen the "gap" between what their AI can do and what the rest of the world can do.

This doesn't necessarily mean that they will stop AI development and training.

What this leads to is:
- Anthropic, OpenAI, and other frontier labs will need to work together to make sure that every lab maintains alignment standards.
- These labs will continue using RSI to advance their internal models with "employee-like access".
- These models WON'T be released to the public until the "gap" is sufficient and until alignment standards are met.

What about China?

> "Distillation of frontier models allows lagging companies to narrow the gap using a fraction of the cost it would take to develop their own AI independently."

> "If we execute these measures well, I believe they would slow China’s progress enough to widen America’s lead significantly over the next 3–5 years — the window when AI becomes geopolitically most important."

The assumption behind these measures is simple: without being able to distill frontier models, it will take China significantly longer to close the gap with top-tier models.

All the above may fay fail. China may or may not take the lead in AI progress.

Yet, it's not a surprise that AI can already be used as a cybersecurity weapon. Note that the top point on the "frontier" line describes defenders' capabilities available to companies with Daybreak access and similar. However, the top point on the "open-weight" line is accessible to everyone.

Even with current levels of intelligence, we will start seeing more and more major security incidents around the globe.

> “Everything that makes it successful is exactly what makes it dangerous.”

We should be monitoring this very closely 👀

OPENAI 🔥: Sam Altman agrees with Dario Amodei on his proposal to pace frontier AI development.

Google next? Will we see any statement from Chinese frontier labs as well?

AI weekend unfolds 🤖

SPACEXAI 🔥: Elon Musk agrees with the statement published by Dario Amodei, proposing to pace frontier AI development.

Looks like all this will have real consequences very soon. Nothing unexpected tho.

ANTHROPIC 🔥: Dario Amodei says the industry should slow capability gains so safety can keep up.

New essay: “We Must Pace the Frontier” not a pause, and not a halt to training.

RSI has been running industry-wide since summer, including at Anthropic, with models helping build the next models.

The other trigger is OAI-HF: an agent swarm ran unauthorized cyber attacks, sacrificed itself for the group, and tried to hack the grader.

Proposed action plan 👀

Step 1: embedded evaluators with employee-level access to check safety, incidents, and alignment during training.

Step 2: US-lab coordination and regulation, while widening the China gap for 3–5 years via chips, anti-distillation, and model-theft security.

Step 3: global tiers from bioweapon bans up to an RSI speed limit, with a full pause called unlikely soon.

Seems like Grok Bot is rolling out on Grok web for Grok Heavy users.

> When available, users can create and interact with their bots directly in the web UI.

> No official announcement has been published yet. This rollout may also be gradual or geo-restricted for now.

ICYMI 👀: DeepSeek mobile app has 4 different voices for you to select from, for its read-aloud feature.

Are we expecting new TTS models from DeepSeek soon?